Energy optimization scheduling method and system based on distributed resources

Abstract

The present invention relates to the technical field of energy scheduling, and discloses an energy optimization scheduling method and system based on distributed resources, including converting the first waste heat into electric energy according to the waste heat parameters of the first waste heat, connecting the electric energy converted from the waste heat to the park smart microgrid, and connecting the second waste heat to the park thermal energy network; converting the first byproduct into combustible gas according to the product parameters of the first byproduct, and selecting the corresponding combustible gas to convert into electric energy according to the gas distribution strategy, connecting the electric energy converted from the gas to the park smart microgrid, and connecting the remaining combustible gas to the park gas energy network; establishing an energy multi-objective scheduling model, and solving the model to obtain the optimal energy distribution strategy; optimizing the energy scheduling according to the energy priority weight and the energy demand of the energy-consuming equipment in the industrial park. The present invention realizes the efficient utilization of industrial waste heat and byproducts, and improves the safety and stability of energy network operation.

Description

Technical Field

[0001] The present invention relates to the technical field of energy scheduling, and in particular to a method and system for optimizing energy scheduling based on distributed resources. Background Art

[0002] In industrial parks, the problems with existing methods for the reuse of industrial waste heat and industrial by-products are that, on the one hand, since there are many different types of enterprises in industrial parks, and the waste heat temperatures generated by different enterprises in the production process vary greatly and the forms of the by-products generated are also diverse, the existing methods cannot uniformly and effectively treat these various types of waste heat and by-products; on the other hand, there is currently no good solution for how to smoothly and efficiently connect the energy converted from industrial waste heat and industrial by-products to the park energy network, and how to minimize energy waste and supply-demand mismatch given the obvious volatility and uncertainty of energy demand within the park. Summary of the invention

[0003] In order to solve the above technical problems, the present invention provides an energy optimization scheduling method and system based on distributed resources, so as to achieve efficient utilization of industrial waste heat and by-products and achieve the technical effect of improving the safety and stability of the operation of the park energy system.

[0004] In a first aspect, the present invention provides an energy optimization scheduling method based on distributed resources, the method comprising:

[0005] According to the waste recycling plan, the industrial waste heat and industrial by-products in the industrial park are divided into primary waste heat, secondary waste heat and primary by-products;

[0006] According to the waste heat parameters of the first waste heat, the first waste heat is converted into electric energy, the electric energy converted from the waste heat is connected to the park smart microgrid, and the second waste heat is connected to the park thermal energy network;

[0007] According to the product parameters of the first by-product, the first by-product is converted into combustible gas, and according to the preset gas distribution strategy, the corresponding combustible gas is selected to be converted into electrical energy, the electrical energy converted from the gas is connected to the park smart microgrid, and the remaining combustible gas is connected to the park gas energy network;

[0008] An energy multi-objective scheduling model is established based on the energy network, and a particle swarm optimization algorithm is used to solve the model to obtain an optimal energy allocation strategy. According to the optimal energy allocation strategy, the energy flow of the energy network is controlled, and the energy network includes the park smart microgrid, the park thermal energy network, and the park gas energy network;

[0009] The energy priority weights are determined based on the cleanliness and conversion efficiency of various energy sources in the controlled energy network, and the energy is optimally scheduled based on the energy priority weights and the energy demand of energy-consuming equipment in the industrial park.

[0010] Furthermore, the step of dividing the industrial waste heat and industrial by-products in the industrial park into first waste heat, second waste heat and first by-products according to the waste recycling plan includes:

[0011] Obtain waste heat parameters of industrial waste heat and product parameters of industrial by-products in the industrial park, calculate the waste heat grade based on the waste heat parameters, and calculate the usable amount of waste heat based on the waste heat medium characteristics and waste heat sources;

[0012] Input waste heat parameters, product parameters and waste heat availability into a pre-established waste comprehensive utilization model to obtain a waste recycling plan;

[0013] According to the waste recycling scheme, industrial waste heat is divided into first waste heat for conversion into electrical energy and second waste heat for heating, and industrial by-products are divided into first by-products for gasification and second by-products for non-energy supply.

[0014] Furthermore, the steps of converting the first waste heat into electric energy according to the waste heat parameters of the first waste heat, connecting the electric energy converted from the waste heat into the park smart microgrid, and connecting the second waste heat into the park thermal energy network include:

[0015] determining whether the waste heat temperature of the first waste heat is higher than a first temperature threshold, if so, converting the waste heat into electrical energy by an organic Rankine cycle, and if lower than the first temperature threshold and higher than a second temperature threshold, converting the waste heat into electrical energy by a Kalina cycle;

[0016] Obtaining electric energy parameters of electric energy converted from waste heat, and judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters; if so, connecting the electric energy converted from waste heat to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements;

[0017] According to the heating parameters and load requirements of the park's thermal energy network, the second waste heat is adjusted to different temperature levels and connected to the park's thermal energy network through regulating valves and variable-speed circulation pumps.

[0018] Furthermore, the steps of converting the first by-product into combustible gas according to the product parameters of the first by-product, selecting the corresponding combustible gas to convert into electrical energy according to the gas distribution plan, and connecting the remaining combustible gas to the park gas energy network include:

[0019] Determine whether the carbon content of the first by-product is higher than a first carbon content threshold value, if higher, use fluidized bed gasification technology to gasify the first by-product into a combustible gas, if lower than the first carbon content threshold value and higher than a second carbon content threshold value, use pyrolysis gasification technology to gasify the first by-product into a combustible gas;

[0020] Analyze the converted combustible gas to obtain the calorific value and composition of the combustible gas, use the model predictive control algorithm to determine the gas distribution strategy, and select the corresponding combustible gas to be transported to the gas turbine for electrical energy conversion based on the gas distribution strategy;

[0021] Obtaining electric energy parameters of electric energy converted from combustible gas, judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters, and if so, connecting the electric energy converted from combustible gas to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements;

[0022] According to the pressure and flow requirements of the park's gas energy network, the remaining combustible gas is pressure-regulated and gas purified, and connected to the park's gas energy network through an intelligent valve control system.

[0023] Furthermore, the steps of establishing an energy multi-objective scheduling model based on the energy network and solving the model using a particle swarm optimization algorithm to obtain an optimal energy allocation strategy include:

[0024] An energy multi-objective scheduling model is established with the optimization objectives of minimizing primary energy consumption, maximizing renewable energy utilization and minimizing energy network operation costs;

[0025] The particle swarm optimization algorithm is used to solve the energy multi-objective scheduling model and obtain the optimal energy allocation strategy.

[0026] Furthermore, the step of optimizing the energy scheduling according to the energy priority weight and the energy demand of the energy-consuming equipment in the industrial park includes:

[0027] According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, a reinforcement learning algorithm or fuzzy reasoning rules is used to determine the energy optimization scheduling strategy;

[0028] According to the energy optimization scheduling strategy, an energy optimization scheduling instruction is generated.

[0029] Furthermore, the steps of using reinforcement learning to determine the energy optimization scheduling strategy include:

[0030] According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, an energy optimization scheduling problem is established with energy cost minimization and energy utilization maximization as objective functions, and energy supply restrictions and equipment operation restrictions as constraints;

[0031] Construct state space based on energy prices, energy demand and energy supply;

[0032] Construct action space according to the start / stop status of energy equipment and load distribution ratio;

[0033] Construct a reward function based on the objective function, constraints, energy demand, and energy priority weights;

[0034] The reinforcement learning algorithm is used to solve the energy optimization scheduling problem and obtain the energy optimization scheduling strategy.

[0035] Furthermore, the steps of using fuzzy reasoning rules to determine the energy optimization scheduling strategy include:

[0036] Formulate fuzzy sets and corresponding fuzzy rules according to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park to obtain a fuzzy rule set;

[0037] Based on the principle of fuzzy logic, fuzzy rules are integrated into the fuzzy reasoning system to obtain the energy optimization scheduling strategy.

[0038] Furthermore, after the step of controlling the energy flow of the energy network according to the optimal energy allocation strategy, the method further includes:

[0039] Acquire historical operation data of the energy network, analyze the historical operation data, and extract network operation characteristic parameters;

[0040] Clustering the pipe network operation characteristic parameters to obtain a normal operation baseline;

[0041] Collect real-time operation data of the energy network, determine the difference between the real-time operation data and a normal operation baseline, and determine whether the energy network is abnormal based on the difference.

[0042] In a second aspect, the present invention provides an energy optimization scheduling system based on distributed resources, the system comprising:

[0043] A reuse analysis module, used for classifying industrial waste heat and industrial by-products in the industrial park into first waste heat, second waste heat and first by-products according to a waste reuse plan;

[0044] an energy network access module, for converting the first waste heat into electric energy according to waste heat parameters of the first waste heat, connecting the electric energy converted from the waste heat to a park smart microgrid, and connecting the second waste heat to a park thermal energy network;

[0045] According to the product parameters of the first by-product, the first by-product is converted into combustible gas, and according to the preset gas distribution strategy, the corresponding combustible gas is selected to be converted into electrical energy, the electrical energy converted from the gas is connected to the park smart microgrid, and the remaining combustible gas is connected to the park gas energy network;

[0046] An energy network scheduling module is used to establish an energy multi-objective scheduling model based on the energy network, and use a particle swarm optimization algorithm to solve the model to obtain an optimal energy allocation strategy. According to the optimal energy allocation strategy, the energy flow of the energy network is controlled. The energy network includes the park smart microgrid, the park thermal energy network, and the park gas energy network.

[0047] The energy optimization scheduling module is used to determine the energy priority weights according to the cleanliness and conversion efficiency of various energy sources in the controlled energy network, and optimize the energy scheduling according to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park.

[0048] The present invention provides an energy optimization scheduling method and system based on distributed resources. The present invention can realize the efficient utilization of industrial waste heat and industrial by-products by converting them into energy. Through energy distribution and optimized scheduling, the converted energy can be smoothly and efficiently connected to the energy network of the park, and the energy network can be operated safely and stably. At the same time, by optimizing and scheduling energy based on energy priority and energy demand, the operating efficiency and energy utilization rate of the energy network of the industrial park can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flow chart of an energy optimization scheduling method based on distributed resources in an embodiment of the present invention;

[0050] Figure 2 It is a structural diagram of an energy optimization scheduling system based on distributed resources in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] See also Figure 1 The first embodiment of the present invention proposes an energy optimization scheduling method based on distributed resources, which includes steps S10 to S50:

[0053] Step S10, according to the waste recycling plan, the industrial waste heat and industrial by-products in the industrial park are divided into first waste heat, second waste heat and first by-products;

[0054] Step S20, converting the first waste heat into electrical energy according to the waste heat parameters of the first waste heat, connecting the electrical energy converted from the waste heat to the park smart microgrid, and connecting the second waste heat to the park thermal energy network;

[0055] Step S30, converting the first by-product into combustible gas according to the product parameters of the first by-product, and selecting the corresponding combustible gas to convert into electric energy according to the preset gas distribution strategy, connecting the electric energy converted from the gas to the park smart microgrid, and connecting the remaining combustible gas to the park gas energy network;

[0056] Step S40, establishing an energy multi-objective scheduling model based on the energy network, and using a particle swarm optimization algorithm to solve the model to obtain an optimal energy allocation strategy, and controlling the energy flow of the energy network according to the optimal energy allocation strategy, wherein the energy network includes the park smart microgrid, the park thermal energy network, and the park gas energy network;

[0057] Step S50, determining energy priority weights according to the cleanliness and conversion efficiency of various energy sources in the controlled energy network, and optimizing energy scheduling according to the energy priority weights and energy demand of energy-consuming equipment in the industrial park.

[0058] The present invention first analyzes the industrial waste heat and industrial by-products generated by various enterprises in the industrial park during the production process to formulate a waste recycling plan for these industrial waste heat and industrial by-products, and classifies the industrial waste heat and industrial by-products according to the waste recycling plan. The specific steps include:

[0059] Obtain waste heat parameters of industrial waste heat and product parameters of industrial by-products in the industrial park, calculate the waste heat grade based on the waste heat parameters, and calculate the usable amount of waste heat based on the waste heat medium characteristics and waste heat sources;

[0060] Input waste heat parameters, product parameters and waste heat availability into a pre-established waste comprehensive utilization model to obtain a waste recycling plan;

[0061] According to the waste recycling scheme, industrial waste heat is divided into first waste heat for conversion into electrical energy and second waste heat for heating, and industrial by-products are divided into first by-products for gasification and second by-products for non-energy supply.

[0062] In this embodiment, in order to improve the recycling rate of industrial waste heat and industrial by-products, a comprehensive waste utilization model is established. The optimal waste recycling scheme is determined by the composition and characteristics of industrial waste heat and industrial by-products. Specifically, for industrial waste heat, thermocouples, ultrasonic flowmeters and pressure sensors are used to obtain waste heat parameters such as waste heat temperature, flow rate and pressure of industrial waste heat in the industrial park. Based on the waste heat parameters, the waste heat grade of industrial waste heat is calculated by thermodynamic simulation method, and the available amount of waste heat is calculated by heat balance method in combination with the waste heat medium characteristics and waste heat sources. For industrial by-products, the types and production of industrial by-products in the park are obtained, corresponding collection devices are set for different types of by-products, the collected by-products are classified and sorted, and the by-product components are detected by X-ray fluorescence spectrometer XRF and gas chromatography-mass spectrometry GC-MS.

[0063] The waste comprehensive utilization model in this embodiment is constructed based on a multi-objective optimization framework, with maximizing economic benefits and minimizing environmental impact as optimization goals, and waste heat temperature, flow rate, by-product types, quantities and components as input variables, and a genetic algorithm is used to solve the problem, thereby obtaining a waste recycling scheme for industrial waste heat and industrial by-products. In the waste recycling scheme, industrial waste heat is divided into a first waste heat that can be used for waste heat power generation and a second waste heat for regional heating, and industrial by-products are divided into a first by-product that can be used for energy conversion, i.e., gasification, and a second by-product for non-energy conversion. For example, for waste heat at 300°C and a flow rate of 200 cubic meters per hour, it is used as the first waste heat for power generation, and for solid waste such as fly ash, it is used as the second by-product for the production of lightweight wallboards, etc.

[0064] In a preferred embodiment, for industrial by-products, the present invention automatically classifies solid, liquid and gas by-products through an intelligent classification and collection device, determines the applicable pretreatment process through database comparison, and uses automated equipment to perform pretreatment operations such as screening, crushing, and dehydration to prepare for subsequent reuse. For example, for sludge with a moisture content higher than 30%, a centrifugal dehydration process is used; for solid waste with a particle size greater than 10 mm, a jaw crusher is used for crushing, and sensors are used to monitor environmental parameters in the by-product storage device, such as temperature and humidity, and to determine in real time whether the storage conditions deviate from the optimal range. When deviated, such as when the temperature exceeds 35°C or the humidity is higher than 60%, the environmental parameters are automatically adjusted, and the by-product storage information is integrated with the waste heat utilization data to provide a basis for comprehensive utilization decisions.

[0065] For industrial waste heat and industrial by-products classified according to the waste recycling plan, energy conversion will continue and the converted energy will be connected to the energy network of the industrial park. The treatment steps for industrial waste heat include:

[0066] determining whether the waste heat temperature of the first waste heat is higher than a first temperature threshold, if so, converting the waste heat into electrical energy by an organic Rankine cycle, and if lower than the first temperature threshold and higher than a second temperature threshold, converting the waste heat into electrical energy by a Kalina cycle;

[0067] Obtaining electric energy parameters of electric energy converted from waste heat, and judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters; if so, connecting the electric energy converted from waste heat to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements;

[0068] According to the heating parameters and load requirements of the park's thermal energy network, the second waste heat is adjusted to different temperature levels and connected to the park's thermal energy network through regulating valves and variable-speed circulation pumps.

[0069] In the present embodiment, a temperature-based electric energy conversion method is adopted for the first waste heat used for electric energy conversion. Specifically, the waste heat temperature of the first waste heat is collected by a temperature sensor to determine whether the waste heat temperature is higher than a preset first temperature threshold; if the waste heat temperature is higher than the first temperature threshold, the organic Rankine cycle is selected for waste heat conversion; if the waste heat temperature is lower than the first temperature threshold and higher than the preset second temperature threshold, the Kalina cycle is adopted for waste heat conversion. For example, the organic Rankine cycle is adopted for waste heat with a waste heat temperature higher than 250 degrees, and the Kalina cycle is adopted for waste heat conversion if the temperature is between 120°C and 250°C. At the same time, corresponding heat exchangers and working fluid circulation systems are respectively provided for waste heat in different temperature ranges. For the organic Rankine cycle, R245fa is selected as the working fluid, and the Kalina cycle uses an ammonia-water mixture as the working fluid. The electric energy conversion method in the present embodiment can effectively improve the efficiency of electric energy conversion.

[0070] For the electric energy converted from waste heat, smart meters and power quality analyzers are used to collect real-time parameters of electric energy, including voltage deviation, frequency fluctuation, and harmonic content. Based on the electric energy parameters, it is determined whether the electric energy meets the grid connection requirements. If the requirements are met, the electric energy is connected to the park's smart microgrid. If not, the fuzzy control-based power quality optimization algorithm is started to adjust the switching of the reactive compensation device and the output parameters of the inverter to make the power quality meet the grid connection standards. For example, the power quality analyzer uses the FFT algorithm to calculate the harmonic content. When the total harmonic distortion rate exceeds 5%, the power quality optimization is triggered. The fuzzy control algorithm automatically adjusts the switching state of the reactive compensation device according to the voltage deviation of ±7% and the frequency fluctuation of ±0.2Hz, and controls the inverter output at the same time to maintain the power factor of the grid connection point above 0.95, thereby meeting the grid connection requirements.

[0071] For the second waste heat used for heating, the temperature and flow of the second waste heat are measured by temperature sensors and turbine flowmeters, and the calorific value of the waste heat is calculated by a heat calculation device based on the thermal enthalpy difference method. According to the heating parameters and load requirements of the park's thermal energy network, a multi-stage plate heat exchanger is used to adjust the waste heat of different temperature levels to suitable ranges. For example, high-temperature waste heat >150°C is used for steam production, medium-temperature waste heat 80-150°C is used for heating, and low-temperature waste heat <80°C is used for preheating or heat pump systems. Finally, the second waste heat is connected to the park's thermal energy network through a regulating valve and a variable-speed circulation pump.

[0072] The treatment steps for industrial by-products include:

[0073] Determine whether the carbon content of the first by-product is higher than a first carbon content threshold value, if higher, use fluidized bed gasification technology to gasify the first by-product into a combustible gas, if lower than the first carbon content threshold value and higher than a second carbon content threshold value, use pyrolysis gasification technology to gasify the first by-product into a combustible gas;

[0074] Analyze the converted combustible gas to obtain the calorific value and composition of the combustible gas, use the model predictive control algorithm to determine the gas distribution strategy, and select the corresponding combustible gas to be transported to the gas turbine for electrical energy conversion based on the gas distribution strategy;

[0075] Obtaining electric energy parameters of electric energy converted from combustible gas, judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters, and if so, connecting the electric energy converted from combustible gas to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements;

[0076] According to the pressure and flow requirements of the park's gas energy network, the remaining combustible gas is pressure-regulated and gas purified, and connected to the park's gas energy network through an intelligent valve control system.

[0077] In this embodiment, the first by-product used for energy conversion is targeted, and no excessive restrictions or explanations are given to the second by-product used for non-energy conversion. For the first by-product, in this embodiment, gasification is used for treatment. Specifically, the by-product is analyzed by X-ray fluorescence spectrometry (XRF) and thermogravimetric analysis (TGA) to determine whether the carbon content of the industrial by-product is higher than a preset first carbon content threshold; if the carbon content is higher than the first carbon content threshold, the fluidized bed gasification technology is used to convert the industrial by-product into a combustible gas; if the carbon content is lower than the first carbon content threshold and higher than the preset second carbon content threshold, the pyrolysis gasification technology is used to convert the industrial by-product into a combustible gas. For example, if the carbon content is greater than 60%, the fluidized bed gasification technology is used to gasify at a temperature of 800-900°C and a pressure of 0.1-0.5MPa. If the carbon content is between 40% and 60%, the pyrolysis gasification technology is used to treat at a temperature of 500-700°C and normal pressure to convert the by-product into a combustible gas. At the same time, the heat generated by the gasification process can also be recovered through a heat exchanger for preheating the feed or supplementing the waste heat utilization system.

[0078] In order to improve the utilization rate of fuel gas, in this embodiment, not all combustible gases are converted into electrical energy. Instead, part of the combustible gases are selected for electrical energy conversion according to the gas distribution strategy. Specifically, a gas analyzer is used to detect the calorific value and composition of the converted combustible gas, and the optimal gas distribution plan is calculated through a model predictive control (MPC) algorithm. The MPC algorithm optimizes gas distribution with a preset duration, such as 15 minutes, and considers factors such as power generation efficiency, pipeline network pressure and user needs. The corresponding combustible gas is selected to be transported to the gas turbine for electrical energy conversion, and the remaining gas is connected to the park gas energy network, thereby improving the comprehensive utilization rate of the fuel gas. In addition, the high-temperature flue gas discharged by the gas turbine can also pass through the waste heat boiler to further recover heat energy, thereby improving energy utilization efficiency.

[0079] For the electric energy converted from combustible gas, the same processing method is adopted as the electric energy converted from waste heat, that is, the real-time parameters of the electric energy, including voltage deviation, frequency fluctuation and harmonic content, are collected by equipment such as smart meters and power quality analyzers. The electric energy parameters are used to determine whether the electric energy meets the grid connection requirements. If the requirements are met, the electric energy is connected to the park's smart microgrid. If not, the fuzzy control-based power quality optimization algorithm is started to adjust the switching of the reactive compensation device and the output parameters of the inverter to make the electric energy quality meet the grid connection standards.

[0080] For combustible gas that is not converted into electricity, according to the pressure and flow requirements of the park's gas energy network, a multi-stage centrifugal compressor or a pressure regulating valve is used to adjust the gas pressure, and the combustible gas is purified by activated carbon adsorption and low-temperature separation devices to remove sulfides and moisture. The gas composition is monitored in real time by a chromatograph to ensure that it meets safety standards, and then the combustible gas is connected to the park's gas energy network through an intelligent valve control system. For example, a four-stage centrifugal compressor is used to pressurize 0.1MPa gas to 0.4MPa, and an activated carbon adsorption tower removes H2S to reduce the content to below 5ppm; the low-temperature separation device removes moisture at -20℃ to a dew point of -40℃. The online chromatograph analyzes the gas composition every 5 minutes to ensure that the CH4 content is >90%, CO2 <3%, etc., to meet the grid connection requirements.

[0081] In the present invention, converting industrial waste heat and industrial by-products into energy and connecting them to the energy network of the industrial park is only the first step of energy optimization scheduling. In this step, the energy conversion plan is determined only based on the characteristics of industrial waste heat and industrial by-products. When these converted energies are connected to the network, they will actually affect the regulation of the original energy network. In order to achieve safe, stable and efficient access of the converted energy to the energy network, in this embodiment, an energy multi-objective scheduling model is established to determine the energy distribution strategy of the energy network. The energy distribution strategy here is mainly used to control the input and output of the energy network. The specific steps include:

[0082] An energy multi-objective scheduling model is established with the optimization objectives of minimizing primary energy consumption, maximizing renewable energy utilization and minimizing energy network operation costs;

[0083] The particle swarm optimization algorithm is used to solve the energy multi-objective scheduling model and obtain the optimal energy allocation strategy.

[0084] In this embodiment, energy flow control of the energy network is achieved based on energy routers and distributed controllers. Specifically, first, real-time data of electricity, heat and gas networks are collected through energy routers. The real-time data include grid load, heat network temperature and pressure, and gas flow parameters. Considering the economy and environmental protection of energy, minimizing primary energy consumption, maximizing renewable energy utilization, and minimizing energy network operation costs are taken as optimization goals. An energy multi-objective scheduling model is established, and the particle swarm algorithm is used to optimize and iterate the model to obtain the optimal energy allocation strategy. According to the strategy, a distributed controller is used to control the input and output of the energy network. Preferably, the distributed controller adopts a redundant design with a control period of 100ms, which can adjust the power balance of the grid, the supply and return water temperature difference of the heat network, and the gas pipeline network pressure in real time, thereby ensuring that the converted energy can be smoothly and efficiently connected to the energy network, and ensuring the safety and stability of the energy network operation.

[0085] The energy network of the industrial park is mainly used to provide energy supply for the energy-consuming equipment of enterprises in the park. Energy demand can be divided into electricity demand, heat demand and gas demand. According to the conventional supply strategy, for electricity demand, priority is given to the use of electricity converted from industrial waste heat and by-products. If the converted electricity is in excess of demand, the surplus electricity will be stored in electrochemical energy storage facilities or flywheel energy storage facilities. If the electricity is in short supply, electricity will be drawn from the power grid or energy storage facilities. For heat demand, priority is given to direct heating using industrial waste heat based on the available amount of waste heat. For gas demand, priority is given to the use of combustible gases converted from industrial by-products.

[0086] In a preferred embodiment, in order to improve the utilization rate of energy, the present invention provides a third step of optimization scheduling based on the above two-step optimization scheduling. The optimization scheduling in this step is based on the energy demand of energy-consuming equipment in the park and the priority of various energy sources, and optimizes the energy network. The specific steps include:

[0087] According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, a reinforcement learning algorithm or fuzzy reasoning rules is used to determine the energy optimization scheduling strategy;

[0088] According to the energy optimization scheduling strategy, an energy optimization scheduling instruction is generated.

[0089] In this embodiment, firstly, the energy priority weight is determined according to the cleanliness and conversion efficiency of the energy. Specifically, the energy priority weight is calculated by the hierarchical analysis method according to the carbon emission factor and primary energy utilization rate of each type of energy. In addition, the energy demand of energy-consuming equipment can be determined based on the demand forecasting model, such as obtaining the historical operation data of the park energy, the historical operation data includes electricity consumption data, heat consumption data, gas consumption data, meteorological data and time characteristic data; using the Pandas library to clean the historical operation data to obtain the cleaned historical operation data; according to the cleaned historical operation data, feature extraction is performed by the sliding time window method to obtain a training data set; based on the training data set, the pre-established demand forecasting model is trained; finally, according to the demand forecasting model, the future electricity demand curve, heat demand curve and gas demand curve of the park are calculated to obtain the energy demand of the energy-consuming equipment. Of course, other methods can also be used to determine the energy demand, which will not be repeated here.

[0090] In this embodiment, based on the energy priority weights and the energy demand of the park's energy-consuming equipment, a method of using reinforcement learning or fuzzy reasoning rules to determine the energy optimization scheduling strategy is provided. The generation methods of these two optimization scheduling strategies are described below.

[0091] When using a reinforcement learning algorithm, the steps include:

[0092] According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, an energy optimization scheduling problem is established with energy cost minimization and energy utilization maximization as objective functions, and energy supply restrictions and equipment operation restrictions as constraints;

[0093] Construct state space based on energy prices, energy demand and energy supply;

[0094] Construct action space according to the start / stop status of energy equipment and load distribution ratio;

[0095] Construct a reward function based on the objective function, constraints, energy demand, and energy priority weights;

[0096] The reinforcement learning algorithm is used to solve the energy optimization scheduling problem and obtain the energy optimization scheduling strategy.

[0097] In this embodiment, the objective function of energy optimization scheduling is constructed with the goal of minimizing energy costs, minimizing environmental pollution, and maximizing energy utilization efficiency. The objective function is converted into a specific mathematical expression according to the energy priority weight and energy demand. At the same time, the energy supply limit and the equipment operation limit are used as constraints to ensure that the scheduling strategy does not violate any constraints while meeting the energy demand. Then, factors such as energy price, energy demand and energy supply are used as state variables to construct the state space, and the scope and resolution of the state space are defined according to the actual situation; according to the available energy types and equipment regulation capabilities in the industrial park, the action space of the scheduling strategy is defined, and the action can include adjusting the energy supply, equipment operation status, etc., that is, the start and stop status of energy equipment and the load distribution ratio; according to the objective function, constraints, energy demand and energy priority weight, a reward function is constructed, such as when the scheduling strategy selects high-priority energy, a higher positive reward is given; otherwise, a lower reward or negative reward is given. When the scheduling strategy can meet the equipment demand, a positive reward is given; otherwise, a negative reward is given, etc. This reward function reflects the performance of the scheduling strategy in many aspects. Finally, reinforcement learning algorithms such as the deep Q algorithm are used to solve the problem and obtain the optimal energy optimization scheduling strategy.

[0098] When fuzzy inference rules are used, the steps include:

[0099] Formulate fuzzy sets and corresponding fuzzy rules according to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park to obtain a fuzzy rule set;

[0100] Based on the principle of fuzzy logic, fuzzy rules are integrated into the fuzzy reasoning system to obtain the energy optimization scheduling strategy.

[0101] In this embodiment, the strategy construction steps based on fuzzy reasoning rules are mainly divided into three steps, namely, determining fuzzy sets, formulating fuzzy strategies, and establishing fuzzy deduction systems. Specifically, fuzzy sets are established based on key factors such as energy type and energy demand, such as "high priority energy", "medium priority energy", "low priority energy", "high demand period", "medium demand period", "low demand period", etc. Then, a series of fuzzy rules are formulated according to energy priority weights and energy demand. For example, "if the current period is a high demand period and the available energy is a high priority energy, then the energy is scheduled first", "if the power generation is much greater than the demand and the energy storage facility is close to full load, then the waste heat power generation power is reduced", etc. Finally, using the fuzzy logic principle, the fuzzy rules are integrated into the fuzzy reasoning system, which can automatically select the optimal energy scheduling strategy according to the real-time energy demand and energy supply.

[0102] The energy scheduling strategy determined by the above steps is converted into control instructions for the corresponding equipment, which realizes the optimal scheduling of energy supply for the industrial park, thereby ensuring the operating efficiency and energy utilization rate of the energy network of the industrial park, and also improving the safety and stability of the energy network operation. It should be noted that the reinforcement learning algorithm or fuzzy reasoning rules used to determine the scheduling strategy in this embodiment can refer to conventional reinforcement learning or fuzzy reasoning methods. The above embodiment is only used as a preference rather than a specific limitation. In addition, other strategy formulation methods can also be used, such as establishing an optimization scheduling model based on the objective function and constraints, and using conventional model solving methods to obtain the optimal solution of the model, which will not be repeated here.

[0103] In a preferred embodiment, in order to further ensure the safety of energy network operation in the industrial park, the present invention also provides a method for abnormal monitoring of the energy network, the specific steps of which include:

[0104] Acquire historical operation data of the energy network, analyze the historical operation data, and extract network operation characteristic parameters;

[0105] Clustering the pipe network operation characteristic parameters to obtain a normal operation baseline;

[0106] Collect real-time operation data of the energy network, determine the difference between the real-time operation data and a normal operation baseline, and determine whether the energy network is abnormal based on the difference.

[0107] In this embodiment, for the energy network in the industrial park, the Internet of Things sensors are deployed to monitor the pipeline network in real time to obtain real-time operation data of the pipeline network, including pressure, flow, temperature, etc. At the same time, based on the historical data of the pipeline network operation, data mining technology is used to establish a pipeline network safety early warning model. By inputting the real-time operation data into the pipeline network safety early warning model, it is determined whether there is any abnormal situation such as leakage or burst in the pipeline network, and an alarm prompt is generated.

[0108] Specifically, the historical operation data of the energy network is preprocessed, and the processed data is feature extracted, and the statistical features such as the mean, variance, and peak value in the sliding time window are calculated to form time series data. The processed time series data is analyzed using the ARIMA model to extract the characteristic parameters of the pipe network operation, including the pressure fluctuation period, flow change trend, and temperature gradient. The characteristic parameters are clustered and analyzed by the K-means clustering algorithm, corresponding to the three states of normal operation, slight abnormality, and severe abnormality, and different modes of pipe network operation are identified to establish a normal operation baseline. A pipe network safety early warning model is established based on the normal operation baseline. The model will judge the difference between the real-time monitoring data and the normal operation baseline. If the pressure drop exceeds the set threshold, the abnormal flow change exceeds the set threshold, or the temperature changes drastically beyond the set threshold, the alarm mechanism is triggered. It should be noted that in this embodiment, only the preferred method of network abnormality monitoring is given. It can also be based on other neural network models such as long short-term memory networks to construct a pipe network safety early warning model to monitor the operation status of the energy network. The specific monitoring method can be flexibly selected according to the actual situation, and there is no excessive restriction here.

[0109] This embodiment provides an energy optimization scheduling method based on distributed resources. The present invention improves the reuse rate of industrial waste heat and industrial by-products through a waste recycling scheme for energy recycling, controls the energy flow of the energy network through an energy distribution strategy, enables the recycled energy to be smoothly and efficiently connected to the energy network, and ensures the safety and stability of the energy network operation after connection. In addition, the present invention optimizes and schedules energy based on energy priority and energy demand, thereby improving the operating efficiency and energy utilization rate of the energy network.

[0110] See also Figure 2 Based on the same inventive concept, a second embodiment of the present invention proposes an energy optimization scheduling system based on distributed resources, including:

[0111] A reuse analysis module 10, for classifying industrial waste heat and industrial by-products in the industrial park into first waste heat, second waste heat and first by-products according to the waste reuse plan;

[0112] The energy network access module 20 is used to convert the first waste heat into electric energy according to the waste heat parameters of the first waste heat, connect the electric energy converted from the waste heat to the park smart microgrid, and connect the second waste heat to the park thermal energy network;

[0113] According to the product parameters of the first by-product, the first by-product is converted into combustible gas, and according to the preset gas distribution strategy, the corresponding combustible gas is selected to be converted into electrical energy, the electrical energy converted from the gas is connected to the park smart microgrid, and the remaining combustible gas is connected to the park gas energy network;

[0114] The energy network scheduling module 30 is used to establish an energy multi-objective scheduling model based on the energy network, and adopt a particle swarm optimization algorithm to solve the model to obtain an optimal energy allocation strategy, and control the energy flow of the energy network according to the optimal energy allocation strategy, wherein the energy network includes the park smart microgrid, the park thermal energy network and the park gas energy network;

[0115] The energy optimization scheduling module 40 is used to determine the energy priority weight according to the cleanliness and conversion efficiency of various energy sources in the controlled energy network, and optimize the energy scheduling according to the energy priority weight and the energy demand of the energy-consuming equipment in the industrial park.

[0116] Furthermore, the reuse analysis module 10 is also used to obtain waste heat parameters of industrial waste heat and product parameters of industrial by-products in the industrial park, calculate the waste heat grade level according to the waste heat parameters, and calculate the waste heat usable amount according to the waste heat medium characteristics and waste heat sources;

[0117] Input waste heat parameters, product parameters and waste heat availability into a pre-established waste comprehensive utilization model to obtain a waste recycling plan;

[0118] According to the waste recycling scheme, industrial waste heat is divided into first waste heat for conversion into electrical energy and second waste heat for heating, and industrial by-products are divided into first by-products for gasification and second by-products for non-energy supply.

[0119] Furthermore, the energy network access module 20 is further used to determine whether the waste heat temperature of the first waste heat is higher than a first temperature threshold, if so, the waste heat is converted into electrical energy by an organic Rankine cycle, and if lower than the first temperature threshold and higher than a second temperature threshold, the waste heat is converted into electrical energy by a Kalina cycle;

[0120] Obtaining electric energy parameters of electric energy converted from waste heat, and judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters; if so, connecting the electric energy converted from waste heat to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements;

[0121] According to the heating parameters and load requirements of the park's thermal energy network, the second waste heat is adjusted to different temperature levels and connected to the park's thermal energy network through regulating valves and variable-speed circulation pumps.

[0122] Furthermore, the energy network access module 20 is further used to determine whether the carbon content of the first by-product is higher than a first carbon content threshold value, if so, the first by-product is gasified into a combustible gas by a fluidized bed gasification technology, and if lower than the first carbon content threshold value and higher than a second carbon content threshold value, the first by-product is gasified into a combustible gas by a pyrolysis gasification technology;

[0123] Analyze the converted combustible gas to obtain the calorific value and composition of the combustible gas, use the model predictive control algorithm to determine the gas distribution strategy, and select the corresponding combustible gas to be transported to the gas turbine for electrical energy conversion based on the gas distribution strategy;

[0124] Obtaining electric energy parameters of electric energy converted from combustible gas, judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters, and if so, connecting the electric energy converted from combustible gas to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements;

[0125] According to the pressure and flow requirements of the park's gas energy network, the remaining combustible gas is pressure-regulated and gas purified, and connected to the park's gas energy network through an intelligent valve control system.

[0126] Furthermore, the energy network scheduling module 30 is also used to establish an energy multi-objective scheduling model with minimization of primary energy consumption, maximization of renewable energy utilization and minimization of energy network operation costs as optimization goals;

[0127] The particle swarm optimization algorithm is used to solve the energy multi-objective scheduling model and obtain the optimal energy allocation strategy.

[0128] Furthermore, the energy optimization scheduling module 40 is also used to determine the energy optimization scheduling strategy by using a reinforcement learning algorithm or a fuzzy reasoning rule according to the energy priority weight and the energy demand of the energy-consuming equipment in the industrial park;

[0129] According to the energy optimization scheduling strategy, an energy optimization scheduling instruction is generated.

[0130] Furthermore, the energy optimization scheduling module 40 is further used to establish an energy optimization scheduling problem based on the energy priority weights and the energy demand of the energy-consuming equipment in the industrial park, with energy cost minimization and energy utilization maximization as objective functions, and energy supply restrictions and equipment operation restrictions as constraints;

[0131] Construct state space based on energy prices, energy demand and energy supply;

[0132] Construct action space according to the start / stop status of energy equipment and load distribution ratio;

[0133] Construct a reward function based on the objective function, constraints, energy demand, and energy priority weights;

[0134] The reinforcement learning algorithm is used to solve the energy optimization scheduling problem and obtain the energy optimization scheduling strategy.

[0135] Furthermore, the energy optimization scheduling module 40 is also used to formulate a fuzzy set and corresponding fuzzy rules according to the energy priority weights and the energy demand of the energy-consuming equipment in the industrial park to obtain a fuzzy rule set;

[0136] Based on the principle of fuzzy logic, fuzzy rules are integrated into the fuzzy reasoning system to obtain the energy optimization scheduling strategy.

[0137] Furthermore, the system is also used to obtain historical operation data of the energy network, analyze the historical operation data, and extract characteristic parameters of the network operation;

[0138] Clustering the pipe network operation characteristic parameters to obtain a normal operation baseline;

[0139] Collect real-time operation data of the energy network, determine the difference between the real-time operation data and a normal operation baseline, and determine whether the energy network is abnormal based on the difference.

[0140] The technical features and technical effects of the energy optimization scheduling system based on distributed resources proposed in the embodiment of the present invention are the same as the method proposed in the embodiment of the present invention, and will not be repeated here. Each module in the above-mentioned energy optimization scheduling system based on distributed resources can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0141] In addition, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0142] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0143] In summary, an energy optimization scheduling method and system based on distributed resources proposed in an embodiment of the present invention divides industrial waste heat and industrial by-products in an industrial park into first waste heat, second waste heat and first by-product according to a waste material recycling scheme; converts the first waste heat into electric energy according to waste heat parameters of the first waste heat, connects the electric energy converted from the waste heat to the park smart microgrid, and connects the second waste heat to the park thermal energy network; converts the first by-product into combustible gas according to the product parameters of the first by-product, and selects the corresponding combustible gas to be converted into electric energy according to a preset gas distribution strategy, and connects the electric energy converted from the gas to the park thermal energy network; The park smart microgrid and the remaining combustible gas are connected to the park gas energy network; an energy multi-objective scheduling model is established based on the energy network, and the particle swarm optimization algorithm is used to solve the model to obtain the optimal energy allocation strategy, and the energy flow of the energy network is controlled according to the optimal energy allocation strategy, and the energy network includes the park smart microgrid, the park thermal energy network and the park gas energy network; according to the cleanliness and conversion efficiency of various energy sources in the controlled energy network, the energy priority weight is determined, and the energy is optimized and scheduled according to the energy priority weight and the energy demand of the energy-consuming equipment in the industrial park. The present invention improves the reuse rate of industrial waste heat and industrial by-products through the waste recycling scheme of energy recycling, controls the energy flow of the energy network through the energy allocation strategy, enables the reused energy to be smoothly and efficiently connected to the energy network and ensures the safety and stability of the operation of the energy network after access, and the present invention optimizes the energy scheduling based on energy priority and energy demand, improves the operation efficiency and energy utilization rate of the energy network.

[0144] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0145] The above-mentioned embodiments only express several preferred implementation modes of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principle of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the protection scope of the claims.

Claims

1. An energy optimization scheduling method based on distributed resources, characterized in that: include: According to the waste recycling plan, the industrial waste heat and industrial by-products in the industrial park are divided into primary waste heat, secondary waste heat and primary by-products; According to the waste heat parameters of the first waste heat, the first waste heat is converted into electric energy, the electric energy converted from the waste heat is connected to the park smart microgrid, and the second waste heat is connected to the park thermal energy network; According to the product parameters of the first by-product, the first by-product is converted into combustible gas, and according to the preset gas distribution strategy, the corresponding combustible gas is selected to be converted into electrical energy, the electrical energy converted from the gas is connected to the park smart microgrid, and the remaining combustible gas is connected to the park gas energy network, specifically including: Determine whether the carbon content of the first by-product is higher than a first carbon content threshold value, if higher, use fluidized bed gasification technology to gasify the first by-product into a combustible gas, if lower than the first carbon content threshold value and higher than a second carbon content threshold value, use pyrolysis gasification technology to gasify the first by-product into a combustible gas; Analyze the converted combustible gas to obtain the calorific value and composition of the combustible gas, use the model predictive control algorithm to determine the gas distribution strategy, and select the corresponding combustible gas to be transported to the gas turbine for electrical energy conversion based on the gas distribution strategy; Obtaining electric energy parameters of electric energy converted from combustible gas, judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters, and if so, connecting the electric energy converted from combustible gas to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements; According to the pressure and flow requirements of the park's gas energy network, the remaining combustible gas is pressure-regulated and gas purified, and connected to the park's gas energy network through an intelligent valve control system; An energy multi-objective scheduling model is established based on the energy network, and the particle swarm optimization algorithm is used to solve the model to obtain the optimal energy allocation strategy. According to the optimal energy allocation strategy, the energy flow of the energy network is controlled. The energy network includes the park smart microgrid, the park thermal energy network and the park gas energy network, specifically including: An energy multi-objective scheduling model is established with the optimization objectives of minimizing primary energy consumption, maximizing renewable energy utilization and minimizing energy network operation costs; The particle swarm optimization algorithm is used to solve the energy multi-objective scheduling model and obtain the optimal energy allocation strategy; According to the cleanliness and conversion efficiency of various energy sources in the controlled energy network, the energy priority weight is determined, and according to the energy priority weight and the energy demand of energy-consuming equipment in the industrial park, the energy is optimized and dispatched, including: According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, a reinforcement learning algorithm or fuzzy reasoning rules is used to determine the energy optimization scheduling strategy; According to the energy optimization scheduling strategy, an energy optimization scheduling instruction is generated.

2. The energy optimization scheduling method based on distributed resources according to claim 1 is characterized in that: The step of dividing the industrial waste heat and industrial by-products in the industrial park into first waste heat, second waste heat and first by-products according to the waste recycling plan includes: Obtain waste heat parameters of industrial waste heat and product parameters of industrial by-products in the industrial park, calculate the waste heat grade based on the waste heat parameters, and calculate the usable amount of waste heat based on the waste heat medium characteristics and waste heat sources; Input waste heat parameters, product parameters and waste heat availability into a pre-established waste comprehensive utilization model to obtain a waste recycling plan; According to the waste recycling scheme, industrial waste heat is divided into first waste heat for conversion into electrical energy and second waste heat for heating, and industrial by-products are divided into first by-products for gasification and second by-products for non-energy supply.

3. The energy optimization scheduling method based on distributed resources according to claim 1 is characterized in that: The steps of converting the first waste heat into electric energy according to the waste heat parameters of the first waste heat, connecting the electric energy converted from the waste heat to the park smart microgrid, and connecting the second waste heat to the park thermal energy network include: determining whether the waste heat temperature of the first waste heat is higher than a first temperature threshold, if so, converting the waste heat into electrical energy by an organic Rankine cycle, and if lower than the first temperature threshold and higher than a second temperature threshold, converting the waste heat into electrical energy by a Kalina cycle; Obtaining electric energy parameters of electric energy converted from waste heat, and judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters; if so, connecting the electric energy converted from waste heat to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements; According to the heating parameters and load requirements of the park's thermal energy network, the second waste heat is adjusted to different temperature levels and connected to the park's thermal energy network through regulating valves and variable-speed circulation pumps.

4. The energy optimization scheduling method based on distributed resources according to claim 1 is characterized in that: The steps of using reinforcement learning algorithm to determine the energy optimization scheduling strategy include: According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, an energy optimization scheduling problem is established with energy cost minimization and energy utilization maximization as objective functions, and energy supply restrictions and equipment operation restrictions as constraints; Construct state space based on energy prices, energy demand and energy supply; Construct action space according to the start / stop status of energy equipment and load distribution ratio; Construct a reward function based on the objective function, constraints, energy demand, and energy priority weights; The reinforcement learning algorithm is used to solve the energy optimization scheduling problem and obtain the energy optimization scheduling strategy.

5. The energy optimization scheduling method based on distributed resources according to claim 1 is characterized in that: The steps of using fuzzy inference rules to determine the energy optimization scheduling strategy include: Formulate fuzzy sets and corresponding fuzzy rules according to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park to obtain a fuzzy rule set; Based on the principle of fuzzy logic, fuzzy rules are integrated into the fuzzy reasoning system to obtain the energy optimization scheduling strategy.

6. The energy optimization scheduling method based on distributed resources according to claim 1 is characterized in that: After the step of controlling the energy flow of the energy network according to the optimal energy distribution strategy, the method further includes: Acquire historical operation data of the energy network, analyze the historical operation data, and extract network operation characteristic parameters; Clustering the pipe network operation characteristic parameters to obtain a normal operation baseline; Collect real-time operation data of the energy network, determine the difference between the real-time operation data and a normal operation baseline, and determine whether the energy network is abnormal based on the difference.

7. An energy optimization scheduling system based on distributed resources, characterized in that: include: A reuse analysis module, used for classifying industrial waste heat and industrial by-products in the industrial park into first waste heat, second waste heat and first by-products according to a waste reuse plan; an energy network access module, for converting the first waste heat into electric energy according to waste heat parameters of the first waste heat, connecting the electric energy converted from the waste heat to a park smart microgrid, and connecting the second waste heat to a park thermal energy network; According to the product parameters of the first by-product, the first by-product is converted into combustible gas, and according to the preset gas distribution strategy, the corresponding combustible gas is selected to be converted into electrical energy, the electrical energy converted from the gas is connected to the park smart microgrid, and the remaining combustible gas is connected to the park gas energy network, specifically including: Determine whether the carbon content of the first by-product is higher than a first carbon content threshold value, if higher, use fluidized bed gasification technology to gasify the first by-product into a combustible gas, if lower than the first carbon content threshold value and higher than a second carbon content threshold value, use pyrolysis gasification technology to gasify the first by-product into a combustible gas; Analyze the converted combustible gas to obtain the calorific value and composition of the combustible gas, use the model predictive control algorithm to determine the gas distribution strategy, and select the corresponding combustible gas to be transported to the gas turbine for electrical energy conversion based on the gas distribution strategy; Obtaining electric energy parameters of electric energy converted from combustible gas, judging whether the electric energy converted from waste heat meets the grid connection requirements according to the electric energy parameters, and if so, connecting the electric energy converted from combustible gas to the park smart microgrid; if not, optimizing the electric energy quality according to the electric energy quality optimization algorithm based on fuzzy control until the electric energy quality meets the grid connection requirements; According to the pressure and flow requirements of the park's gas energy network, the remaining combustible gas is pressure-regulated and gas purified, and connected to the park's gas energy network through an intelligent valve control system; The energy network scheduling module is used to establish an energy multi-objective scheduling model based on the energy network, and use the particle swarm optimization algorithm to solve the model to obtain the optimal energy allocation strategy. According to the optimal energy allocation strategy, the energy flow of the energy network is controlled. The energy network includes the park smart microgrid, the park thermal energy network and the park gas energy network, and specifically includes: An energy multi-objective scheduling model is established with the optimization objectives of minimizing primary energy consumption, maximizing renewable energy utilization and minimizing energy network operation costs; The particle swarm optimization algorithm is used to solve the energy multi-objective scheduling model and obtain the optimal energy allocation strategy; The energy optimization scheduling module is used to determine the energy priority weights according to the cleanliness and conversion efficiency of various energy sources in the controlled energy network, and optimize the energy scheduling according to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, including: According to the energy priority weights and the energy demand of energy-consuming equipment in the industrial park, a reinforcement learning algorithm or fuzzy reasoning rules is used to determine the energy optimization scheduling strategy; According to the energy optimization scheduling strategy, an energy optimization scheduling instruction is generated.

Images