Hello, everyone. Welcome to our Q2 result conference. As usual, we will have our IR, Ronnie, to record the numbers, the financial numbers of Q2 and the first half. Then we will have the Q&A session. Please feel free to just raise your questions if you have any, after the presentation. Thank you for your attendance today. We haven't really have so many people with us, I mean, for the result conference. Thanks for coming. Before the presentation of financial numbers, we will have our CSO, Chief Sustainability Officer, to share our plan of SBT Net-Zero before the financial numbers. The climate issue has been a really, really important topic for every country.
I believe that every company has their own plan toward this, toward the goal of Net Zero. But some of them, they may be questioned about the off flush. For this part, we will also share what we really do and what we really work on it with you. Let's just have the presentation from CSO first. We very recently, we just published our annual sustainability report just last week on our official website. We have already passed the SBT Net-Zero validation at the end of last year. We also announced our climate transition plan, and we are committed to the climate transition toward 2050 SBT Net-Zero. We actually, initially, we have, we have submitted our SBT plan and SBT target in 2017.
We were committed to achieve like, to reduce like 56.6% of our carbon emission by 2025. In reality, we actually reached and achieved this goal. Five, sorry, four years ahead of our schedule. That's why, last year, in 2022, again, we submitted our new target to SBTi and also got the validation, official validation from SBTi. Our new plan, for the next phase, is we are going to. We are targeting to reduce like 90% of our, I mean, carbon emissions, I mean, in Scope 1 and Scope 2. Then we also aim to reduce another 25% of our carbon emission in Scope 3 by 2030. That would be our short-term target.
For the long-term target, which is in 2050, we actually plan to reduce like 90% of the carbon emissions, I mean, in Scope 1 and Scope 2 and Scope 3. That is actually in line with the SBT methodology, which allows all the, I mean, the organizations, enterprises, to reduce at least, I mean, 90% of their carbon emissions. For the very last 10% of the carbon emissions, it is allowed to use other approaches or other methods just as large, such as, using or the carbon sink or like the carbon credits to achieve the final 10% and the last mile of this net-zero.
Therefore, our climate transition plan is actually aligned with the SBT methodology and aim to control the temperature increase, I mean, global warming within 1.5 degrees Celsius. As you can see from the from the figure, in 2022, we have already successfully reduced our market-based carbon emissions by 13.5%. In terms of our achievement rate of the renewable energy adoption. As you can see from the graph, in 2021, the adoption rate of renewable energy within our global operation sites was about 55%. In 2022, we targeted to achieve like 60% of the renewable energy usage, but eventually we actually achieved 63%, which was also ahead of our schedule.
As mentioned at the beginning, we actually take it very seriously to avoid any possibility of any forms of greenwash. What we do here is our plan is completely aligned with the United Nations Integrity Matters Report. We have been always watch this very closely to avoid any possibility of greenwash or over-announcement. Here we also provide this table for your reference. I think that we are doing okay here. As usual, our financial numbers are reported based on IFRS, and all the consolidating numbers have been reviewed by CPA. The Q2 revenue was NT dollars 100.5 billion, representing a 12% year-on-year growth and an 8% sequential increase, which was seasonal.
Gross Profit in Q2 grew by 11% year-on-year and 15% quarter-on-quarter. Shipping margin in Q2 improved to 29.2% from 27.5% in Q1. Thanks to the better scale, but I mean, that was mainly related to the better scale. The GP margin in Q2 slightly dropped from 29.4% a year ago, mainly because of the unfavorable product mix this year. Q2 expense was up 14% year-on-year and 11% quarter-on-quarter. As a result of more business travels and wage increases for the engineers, especially. R&D expense in Q2 grew slightly faster than the SG&A. R&D as a percentage of sales, was 8.8% versus 8.3% in Q1, and 8.5% a year ago.
SG&A, as a percentage of sales, remained at 10.2% compared to Q1, but slightly expanded from 10.1% a year ago. The Q2 OpEx ratio increased to 19.0% versus 18.5% in Q1, and 18.7% a year ago. The Q2 operating profit was up 7% year-on-year, and 22% quarter-on-quarter. Q OP margin also improved to 10.2% versus 9.1% in Q1, but dropped from 10.7% a year ago, due to the lower GP margin and the higher OpEx ratio. The sales-wise, we actually found the fastest growth from power electronics, mainly related to the strong expansion in the EV component business and the healthy demand for power supply business as well.
Followed by the automation segment, where IA moderate, moderately improved from a low base. Building automation decelerated compared to last year. Within infrastructure, the EV charger business was strong. The telecom power business was sluggish, sluggish due to the weakness in the underlying market. Earning-wise, thanks to the improvement in the EV component business, we found a substantial year-on-year expansion for power electronics. Some profit contractions in automation and infrastructure, owing to the profit declines in IA and ICT business. Q2 non-operating profit was NT dollars 1.6 billion, which was similar to Q1. In Q2, we had NT dollars 11.9 billion profit before tax, up 10% year-on-year, and 20% quarter-on-quarter. Here, we also provide the EBITDA numbers for your reference.
Q2 tax expense was about NT$ 2.4 billion, representing a 20% effective tax rate. Q2 net profit after tax was about NT$ 8.1 billion, up 7% year-on-year, and 18% quarter-on-quarter. The Q2 EPS was 3.14. We will have a look at the accumulated numbers of the first half. The first half revenue was NT$ 193.5 billion, up 12% from a year ago. Gross profit in the first half was up 12% year-on-year as well, with a GP margin of 28.4%, same as a year ago. The operating expenses in the first half was up 14% year-on-year, where the SG&A up 14% and R&D up 13%.
The OpEx ratio moderately increased to 18.7% in the first half from 18.5% a year ago. The SG&A's percentage of sales grew to 10.2% from 10.10% a year ago, while the R&D expense ratio also increased to 8.5% from 8.4%. The operating profit was up 9% year-on-year, while the operating margin slightly moderated to 9.7% from 9.9%. Segment-wise, we found a decent sales growth almost across the board, with the power electronics and automation growing faster than infrastructure. Profit-wise, we saw a significant earnings expansion in power electronics and had some profit contractions in automation and infrastructure.
Thanks to the improvement in the investment gains and interest income, the non-operating profit was NT$3.2 billion versus NT$2.2 billion a year ago. In first half, we had NT$21.9 billion pre-tax income, up 13% from a year ago. In the first half, the tax expense was about NT$4.1 billion, representing a 20% effective rate. The net profit after tax was NT$15.1 billion versus 13.4, NT$13.7 billion a year ago. The EPS in the first half was 5.80, up 10% from a year ago.
As for the question, I mean, for the first question from the audience, after many automakers have announced their adoption of charging standard of Tesla for the North American market, is likely to have a significant impact on the industry's development. This adoption indicates a shift towards a standardized charging infrastructure. Is it possible that the market will slow down temporarily due to the product spec modifications and resulting in the short-term shipping gap? What kind of the impact would it have on your business? For this question, I actually also consulted our engineers. For the CCS standard and non-standards, the only difference is actually at the connector. There is actually not a really high barrier in terms of technology or the product design.
Because Tesla actually announced that they will open their IPs to everybody outside of its ecosystem. So far, we don't see a significant impact on our business, I mean, as a supplier. I do believe that there is going to be some consolidations, I mean, some consolidation in the market, and that is actually also good for the market development. Maybe in the near term, we will see some, some of the customers, they might be more hesitant because they are still going to observe this trend. For the long run, we do believe that this kind of consolidation is actually good for the overall market development. The next question is about the AI servers. During the shareholders meeting, it was mentioned that there are many opportunities in the AI server power supply business.
Can you give us more colors about the orders, the outlook for the next year and the year after? Before answering this question, I would actually like to ask the question: How we can really define the definitions of AI servers? As far as I know, if I, I mean, go to ask, like, three engineers, like each one of them, they may just give me or give us, like, different definitions of the AI servers. You may define the AI servers are those servers with the graphic cards and with GPU or a specific GPU.
For our power supplies, I mean, we actually have been shipping to many data center clients, but To be honest, we don't really know the, how they really, I mean, utilize these products, like, either they are for the AI servers or are not for the AI servers. I do believe that this, I mean, the development of this AI server or this whole AI application is still in the very early stage of its development. Currently, the AI server, as a percentage of total number in the server market, still holds a very small share now. The business models of AI operators are not yet well-defined. The trend, I do believe, is going to be more volatile rather than a straight line up.
I think some people may also concern about the development of AI technology will going to take over some of the job opportunities from human beings. Personally, I do believe the AI development or the AI applications is going to create more job opportunities than take, I mean, taking over the job opportunities from human beings. It will also increase our product productivity at the same time. We are really, we are, we are kind of optimistic about this trend. As in specs and ASP or the wattage density of AI servers, I think generally speaking, they are all higher than the conventional powers.
Because the specs and requirements are for the AI servers are all much higher than the conventional servers, so supposedly the margins and ASP of AI servers of our products are also higher than the conventional server. Could you please share your view on the growth and the profitability of the EV component business this year? Has it already exceeded the company's initial expectations? I think in the first half of the year, revenue from the EV component business has largely matched, exceeded our, I mean, our expectations. For the second half of the year, we do believe the customer demand has remained quite strong. We shall continue to sustain a healthy growth for this business.
As for the business opportunities for AI server cooling, have you benefited from it? Can you share with us your progress on this technology or this business? As a major player in data center cooling solutions, we do have many projects in hand. Currently, even though, there are various new cooling technologies, I mean, in the market, each one of them actually has its own advantages and disadvantages, making the direction less clear in the short term. On the other hand, we also offer, I mean, various products to customers, not only the liquid cooling solutions, but also the traditional air cooling, as well as the single-phase or two-phase liquid cooling solutions. The overall server market seems to see weaker demand in the first half of the year.
How is the business in this area? What are the expectations for the market sector? Year-to-date, the growth momentum of our traditional server, of the traditional server market has indeed slowed down compared to 2022. One of the main reasons could be the backlog of orders from 2021 have been largely fulfilled last year. For the long run, for the long run, we do believe that this business is definitely, is definitely, I mean, will continue to grow. Just for the short term, I think it's actually going to gradually bottoming up, bottoming up for the overall server market this year. We might see some modest recovery next year for the overall server market. Okay, is there any sign of recovery in the China IA market?
I think in the first half of the year, the China IA market seems to have shown some signs of slow recovery. Initially, it was expected that the easing of post-pandemic restrictions will lead to increased demand. In reality, the overall business environment is still adjusting. There hasn't been a significant improvement in the industrial demand. Our business in this sector grew by 13% in the first half of the year, which is in line with the performance of our competitors. Despite the fact that the second half of the year may have a lower comparison base, making it challenging to anticipate an acceleration in this one. How do you see the consumer market?
I think the market, the overall market, has been gradually entering to the late stage of the whole de-stocking process. We don't really expect, I mean, the consumer market is going to see any kind of, I mean, significant recovery, but I think it's actually going to stabilize pretty soon. Your one of your main competitor in the Power Supply business recently just announced their results of Q2, and they actually improved their margins and profitability this year. Do you have any comment? What do you, I mean, what's your view on your own profitability? We don't comment our competitors' performance, but we are happy for them. For our own GP margin, I think in the long run, we always expect to at least maintain our 30% level, GP margin level.
Could you please share about the details of your CapEx plan for this year and for the next few years? We are actually building up building our, some of our office buildings and expanding our production lines in different areas. For example, we not only have new construction plans in Taiwan, but also in China and Europe. We are also planning to have another two new factories in Thailand. We also the going to complete it going to complete the construction plans constructions for the new factories in India. The capacity for the India, from the India factories are mainly for the domestic market in the initial stage.
Can you please, I mean, share us a rough idea about like what the average wattage of AI servers will be? From the previous conference call, you actually mentioned because the EV component business is going to outgrow other businesses this year. That actually puts some pressures on the GP margins. Considering that the AI servers or the server powers for data centers actually has also been growing this year, is that going to help on your overall GP margins? I think in terms of the average wattage of AI servers, is really subject to the architectures of different data centers.
In the old days, the power outage of servers were normally within the 1,000-2,000, 1,000-2,000 watts range, the range of 1,000-2,000 watts. Nowadays, I think many of them, many of the, sorry, the power outage of the power, power wattage of many AI servers or the data center servers, actually within the range of 3,000-4,000 or even like 6,000 watts per unit. For GP margin, I mean, of our EV component business, I think our strategy is very simple. We would like to enlarge our scale to improve our profitability, the overall profitability of this, for this business.
I can be sure that, even in the long run, for this business, in terms of the GP margin, it is never going to be like 30%, 40% or like higher than other businesses. There is still room for us to have some, to see some upside potential for the profitability for this business. Currently, the overall server power supplies actually accounts for like nearly half of our power supply business. My question will be related to the telecom power business. As far as I know, this business, I mean, in terms of the market demand, has been pretty weak in these two years, and then it even becomes a drag on your margins, OP margin.
Do you have any plan or any kind of any approaches, strategies to improve this kind of situation for the telecom power business? I think for the telecom power business, this market is actually the nature of this market is pretty cyclical. It's not like, okay, this business or this market is going to be saturated or to be more commoditized. It's just the nature of it, it's just more cyclical. For the second question related to the EV business. Strategically, we actually tend to work with the European or the Western OEMs in terms of the clientele. Many of our clients, they actually also have joint ventures, I mean, in China. Currently, we don't really have many China OEMs as our customer.
I think the main reason is the competition there is much more fierce, especially the market itself and also the customers, they are more price sensitive by nature. Our advantage is, has always been on the product development and the technology. If the clients are more price sensitive, it's actually hard for us to really differentiate ourself, ourselves and then also to protect our margins. That's the reason. Okay, the next question is: Do you have any plan to further increase your shareholding in Delta Thailand? No, we don't have a plan. We don't have that plan now. Some people said that you have the penetration into the one of the major Japanese OEMs. Could you please share the details of of it? Sorry, we don't comment on any specific clients.
Could you please share your view on the profitability of your EV component business, given its margin is lower than your corporate average? I think even though by nature, the margin of the EV business is going to be dilutive to the overall GP margin . On the bottom line level, it, when we see the, I mean, much better scale and much better and much more sales contributions coming from the EV business, it's still going to be very beneficial to the bottom line and to the absolute, absolute profits. Can you share your plans on the capacity, the new capacity, I mean, Delta Thailand? I think some of them, or like, a big part of it, is actually for the EV business.
Because for many of the EV component business, we are actually the sole source, I mean, to the clients, so we got to be prepared for that. Can you please share your view on your automation segment, both for the industrial automation business and building automation business? I think that our idea for the, I mean, by nature, these two business or these two markets are quite different. In terms of the product design or, for the industrial automation market, from product to product, from component to component, for different sector of different verticals could be variant. For the building automation, the products or components are more generic. The key is more about the integration of different products, and the on the software side as well.
I think for building automation, as there is actually growing consciousness about the energy efficiency of the buildings. We also see some of, I mean, some many countries, their governments have already, already set up some new policies regarding the energy efficiency of the buildings, of the new buildings. The improvement of the energy efficiency is actually our expertise. That's the reason why we have this focus on the building automation business. I do believe that we will have many, many business opportunities in our infrastructure business, infrastructure segment. Not only on the data center solution side, but also on the energy infrastructure segment, including our Energy Storage System business. Our EV Charging business as well.
As there are going to be more and more electric vehicles on the road, the importance of the development and constructions of microgrid has also growing substantially in order to deal with this kind of, I mean, challenging. Can you also please tell us, or give us some idea about your energy storage system business? As we actually explained before, it's not going to work. I mean, for a new renewable energy or the microgrid infrastructure, it's not going to work without the help of energy storage systems. The energy storage systems can actually not only, I mean, can help to save the energy during the daytime, for example, I mean, from the solar systems, but also it can quickly or rapidly charge and discharge from the grid.
It's not going to work, I mean, for, without the help of the energy infrastructure, Energy Storage System. In Taiwan, in 2024, there is going to be a new policy. I mean, in terms of the carbon fees. I think currently, as we just, I mean, elaborated, at the beginning of this meeting, because we have already use, I mean, like o r the, the, the renewable energy has already accounted for 63% of our total electricity usage. For this year, that we actually target to achieve maybe 80% of the renewable energy usage. I think it's not, it's not going to have, I mean, any kind of, I mean, substantial impact on Delta.
Of course, we would also like to help the supply chain to moving toward this renewable energy trend and also moving toward to the carbon neutrality. Now we do believe that this is our responsibility. As you can see from the, the, the, the, the presentation, I actually, at the beginning of this presentation, I actually mentioned the carbon emissions in Scope 1 and Scope 2 and Scope 3. For the Scope 1 and Scope 2, that is mainly related to the carbon emissions within the enterprise or corporate itself. For the Scope 3, it's more related to the value chains. It's not just, I mean, our responsibility to, to, to, to help our suppliers or customers to reduce their carbon emissions.
In order to achieve our carbon emission target for the Scope 3, we will also need to do that. That would be one of our major focuses in the next stage, because we have already done quite a lot for the Scope 1 and Scope 2. The last question will be: What will be your growth drivers in the future? For the growth drivers, I think the EV business is definitely going to be the growth driver for the near term and to, for the midterm. For the data center and energy infrastructure, as I said, I do tend to believe that they are going to be our mid to long-term growth drivers in the future. For the, within the automation segment, I think the industrial automation business is going to be more steady growing.
Even for our component businesses, such as our Cooling Fan business and our Power Supply business, as we also have, I mean, the penetration into many new applications and also to the growing areas such as data center. We do believe that they are going to have pretty decent growth in the mid to long term. Do you have any idea about your market share in terms of your power supplies of in the AI server market? I think, as I said at the beginning, because the definition of AI server is actually pretty vague, so it's hard to really calculate the market share of the product. For the AI server, my question is still re, related on the AI servers. Do you have any specific numbers, just like.
Okay, so how, so what is the growth rate will you expect for this AI server, server Power Supply, for the next year? I know that everyone actually is highly interested in the AI server market. Still, I think the AI operators will still need to figure out the business models. I think we are actually just in a very initial stage, I mean, of this AI, AI development. At least from my point of view, there is still isn't, there still isn't any kind of mature applications for the AIs. As I mentioned, I think despite the fact that many hyperscalers have been aggressively investing into the AI server deployment, but still, because this market is still evolving.
I think it's really hard for anyone to calculate the market share or to calculate the aggregate number of AI servers. For the EV trend, it's actually more clear. Because the trend in the market is actually to integrate like a few components together, like maybe three-in-one or four-in-one. The advantages and strength, the strengths of, I mean, integrate many components within one bus, is we can make the products much lighter and much smaller. Of course, the technology, I mean, barrier, technological barrier is much higher than making discrete components. Thank you for coming today. All the best.