Scenario description
The image is AI-generated
Energy consumption optimization is directly related to the achievement of the "dual carbon goals" in the "14th Five-Year Plan and 2035 Long-term Goals", and has become the top priority for the development of process manufacturing enterprises. According to the Carbon Emission Ranking of China's Listed Companies (2021), the 100 high-carbon emission companies listed on the list are distributed in eight high-energy-consuming industries: petrochemical, chemical, building materials (cement), steel, nonferrous metals, papermaking, electric power, and aviation. Six of the eight key high-energy-consumption industries belong to the "big manufacturing industry". The high-end, intelligent and green of the traditional manufacturing industry has improved the development requirements of the manufacturing industry for new technologies.
The cement industry has high coal and electricity consumption, and the "two grinding and one burning" process of cement (raw mill, rotary kiln and cement grinding) is the main factor to ensure the stability of cement quality.
solution
Participate in the selection of the Innovation Scenario 50
In 2021, Conch Cement, as the largest cement clinker supplier in Asia, had a keen insight into the fact that the traditional APC (Production Optimization System) iteration capability could not keep up with business needs, the software adaptability was not strong, and the digital precipitation and reuse were limited, resulting in poor global optimization capabilities.
Step 1: Data collection and cleaning. Combined with the experience of process experts, the cement industry brain first extracts massive data from production systems, control systems, equipment management systems, and energy management systems, including quality inspection data, DCS data, environmental data, etc. At the same time, the data is cleaned, noisy data or invalid data is eliminated, and the missing data is supplemented, so as to provide high-quality data assets for the next step of model training.
Step 2: Model building. Advanced machine learning algorithms, neural network algorithms, combined with advanced process control models, model the collected multi-dimensional data, and truly restore the actual production process on the cement production line. And the parameters of the big data model are adjusted to realize the nonlinear mapping relationship from the input parameters to the output parameters.
Step 3: Machine learning. By collecting six months of historical data, analyzing the coupling relationship between up to 100 variables, and predicting the output of the model, the optimal combination range of wind, coal and material can be quantified and visualized, so as to achieve the best clinker quality for the same yield. The highest yield is achieved under the same quality; Or in the case of homogeneous and homogeneous production, the energy consumption is the lowest.
Step 4: Online control. The setting of the process parameters of the final production line will be combined with the range of process parameters, step size information, real-time values of process parameters, etc., and the cement industry brain will conduct multi-variable comprehensive analysis, and optimize the output, quality, and energy consumption of each working condition in real time, and recommend a set of optimal process parameters to be rewritten back to the decentralized control system in real time, so as to realize the automatic driving and unattended operation of the cement core production process.
effect
In just two months, Conch Cement achieved highly automatic control of the cement process optimization. In the process, energy consumption is reduced by 2%-3%, which is significant for a plant with a daily cement output of 12,000 tons. In a later period, the energy-saving test of global optimization was gradually completed. The results show that the energy saving level of the system exceeds that of the same type of software from well-known foreign manufacturers.
"About Innovation Scenario 50"
The scene is not a case, it is more precise and abstract. Digitalization is the continuous superposition and iteration of innovation scenarios.
In this context, Titanium Media launched the "Innovation Scenario 50" selection, which selects and interprets 50 innovative scenarios and solutions that are deeply integrated with the whole industry and business every year, and awards and in-depth exchanges are presented at the annual ITValue Summit Digital Value Annual Conference of Titanium Media.
At present, the scene is being solicited, more accurate interpretation, wider exposure, and stronger brand potential, you are welcome to ask questions, and you are more welcome to leave solutions and tools. Click here to post more scene information
Click here to learn more at the 2024 ITValue Summit Digital Value.