Continuous production type industrial plant hybrid energy storage system, dispatching method and storage medium

Abstract

This invention relates to the field of industrial integrated energy management technology, specifically to a hybrid energy storage system, an efficient collaborative scheduling method, and a storage medium suitable for continuous production industrial plants. The system comprises an information sensing and supply layer, a hybrid energy storage and conversion layer, and an optimization decision and control layer connected sequentially. The information sensing and supply layer acquires multi-dimensional information data and multi-energy flow supply data. The hybrid energy storage and conversion layer realizes multi-grade thermal energy storage, electrochemical energy storage, and thermoelectric coupling conversion. The optimization decision and control layer executes scheduling optimization algorithms and includes a heat load prediction module, a multi-source collaborative optimization model module, and a working mode switching module. This invention, by constructing a three-layer architecture and based on MES production scheduling, weather forecasts, and waste heat data, forms a production plan-driven thermal energy storage scheduling method, realizing a shift from passive energy supply to active process matching, significantly improving the planning and accuracy of energy supply.

Description

Technical Field

[0001] This invention relates to the field of industrial integrated energy management technology, specifically to a hybrid energy storage system, an efficient collaborative scheduling method, and a storage medium suitable for continuous production industrial plants. Background Technology

[0002] In continuous industrial production scenarios, the stability and efficiency of the energy system are crucial for ensuring product quality and production safety. Currently, most factory energy solutions integrating photovoltaics and energy storage have significant shortcomings: First, their scheduling logic often revolves around grid electricity prices or simple load shaving, resulting in a severe disconnect from the factory's predictable baseload energy demand based on production plans. Second, the factory's abundant multi-grade waste heat and thermal storage systems lack effective coupling, making it impossible to classify, recover, and store waste heat of different grades. This leads to the downgrading or direct waste of high-quality waste heat, resulting in low energy cascade utilization and overall low efficiency. Electrochemical energy storage is often inappropriately used for large-capacity, long-term energy transport, leading to a sharp decline in cycle life under continuous high-load conditions and poor economic efficiency. Finally, existing emergency solutions primarily focus on power supply during grid failures, failing to fully consider the extreme importance of heat supply in continuous production processes. When unplanned production fluctuations or extreme weather causes sudden changes in heat demand, the lack of a rapid and targeted emergency thermal dispatch mechanism affects production stability.

[0003] Therefore, there is an urgent need for a new solution that can be deeply integrated with the production process, realize intelligent division of labor in electric thermal energy storage, and ensure extremely high reliability. Summary of the Invention

[0004] The purpose of this invention is to provide a hybrid energy storage system, an efficient collaborative scheduling method, and a storage medium suitable for continuous production industrial plants, in order to solve the technical problems in the prior art, such as the disconnect between energy storage scheduling and production planning, the inability to coordinate the use of multiple energy sources, the large lifespan loss of energy storage equipment in continuous production scenarios, and the lack of a flexible guarantee mechanism that takes into account both economic efficiency and production continuity.

[0005] To achieve the above objectives, embodiments of the present invention provide a hybrid energy storage system for a continuous production industrial plant, comprising: The system is sequentially connected to the information sensing and supply layer, the hybrid energy storage and conversion layer, and the optimization decision-making and control layer; among them... The information sensing and supply layer is used to acquire multi-dimensional information data and multi-energy flow supply data; The hybrid energy storage and conversion layer is used to realize multi-grade thermal energy storage, electrochemical energy storage and thermoelectric coupling conversion; The optimization decision and control layer is used to execute the scheduling optimization algorithm; the optimization decision and control layer includes a heat load prediction module, a multi-source collaborative optimization model module, and a working mode switching module; the heat load prediction module is used to predict the optimal heat release power plan of each thermal storage unit based on the data obtained by the information perception and supply layer; the multi-source collaborative optimization model module is used to solve the optimal spatiotemporal matching scheme between multi-grade waste heat recovery, graded storage, and on-demand supply; the working mode switching module is used to evaluate the real-time heating capacity and switch the working mode of the energy storage system according to the process heat demand prediction data.

[0006] Optionally, the heat load prediction module includes a heat load prediction model and a heat release plan optimization model for the thermal storage subsystem, wherein, The heat load forecasting model is used to obtain the baseline heat load forecast values ​​and their confidence intervals for each future period, so as to obtain the baseline heat load demand curve. The heat load forecasting model includes static and dynamic terms. The static term is a physical model based on the production plan, and the dynamic term is a dynamic heat load deviation forecast based on the LSTM model. The thermal storage subsystem heat release plan optimization model is used to obtain the optimal heat release power plan for each thermal storage unit. The thermal storage subsystem heat release plan optimization model includes a cost minimization objective function and constraints. The constraints include thermal balance constraints, dynamic energy balance constraints of thermal storage units, capacity and charge / release power limits of thermal storage units, temperature limits of thermal storage units, and process temperature matching requirements.

[0007] Optionally, the cost minimization objective function is: ; in, This is the unit power consumption cost coefficient for the circulating pump. In order to be in Time period The operating power of the circulation pump of each thermal storage unit The cost coefficient per unit heat loss. In order to be in Time period Heat loss of each thermal storage unit.

[0008] Optionally, the multi-source collaborative optimization model module includes a multi-source collaborative optimization model with the objective of maximizing the overall comprehensive benefit of the system, and the objective function of the multi-source collaborative optimization model is: ; in, For thermal energy value, In order to be in Time period The actual heat demand satisfied at each heat load node. In order to be in Time period Start-up and shutdown status variables of a waste heat recovery device For the first The cost of starting and stopping a single waste heat recovery device In order to be in Time period The recovery power of each waste heat recovery device In order to be in Time period Heat loss of each thermal storage unit For the first Unit power operating cost of a waste heat recovery device The cost coefficient per unit heat loss. This represents the total number of time periods within the scheduling cycle.

[0009] Optionally, the constraints of the multi-source collaborative optimization model include grade matching and heat flow balance constraints, recovery power allocation and logical constraints, thermal storage unit dynamic and capacity constraints, and spatiotemporal transmission losses; wherein, The constraints of quality matching and heat flow balance are: ; in, Indicates in Time period The thermal storage unit sends to the first The effective heat supply released by each heat load node For the first The current available temperature rating of each thermal storage unit. For the first The minimum supply temperature required by each heat load node. For the first The supply temperature level of a direct heat source, The heat transport efficiency coefficient of a direct heat source. In order to be in Time period The first direct heat source to the first The original heat power delivered by each heat load node Indicates in Time period Total heat demand of each heat load node.

[0010] Optionally, the working mode switching module includes a real-time monitoring and prediction module and a threshold judgment module; the real-time monitoring and prediction module is used to evaluate the real-time heating capacity and calculate the heat gap; the threshold judgment module is used to determine the working model to be used in the current cycle based on the heat gap.

[0011] Optionally, the judgment logic of the threshold judgment module includes: Determine whether the heat deficit is greater than the warning threshold; If the heat deficit is determined to be greater than the warning threshold, a warning state is triggered, the equipment status is checked, and emergency logic is prepared to be started. Determine whether the heat deficit is greater than the emergency trigger threshold; If the heat gap is determined to be greater than the emergency trigger threshold, the energy storage system is controlled to switch from normal mode to emergency mode; in the emergency mode, battery protection mode switching, heat storage strategy reconstruction and gap filling calculation operations are performed. Determine if the heat deficit has been eliminated; If the heat deficit is not eliminated, the emergency mode is maintained. If the heat deficit is determined to have been eliminated, the energy storage system is switched to normal mode.

[0012] Optionally, the battery protection mode switching includes: A collaborative optimization model for lifetime cost is constructed, and the objective function of the collaborative optimization model for lifetime cost is: ; in, The constraints of the collaborative optimization model for life cost include system power balance constraints and battery self-constraints.

[0013] On the other hand, the present invention also provides a hybrid energy storage scheduling method for a continuous production industrial plant, wherein the hybrid energy storage scheduling method for the continuous production industrial plant applies any of the above-described hybrid energy storage systems for continuous production industrial plants, and the hybrid energy storage scheduling method for the continuous production industrial plant includes: Acquire multi-dimensional information data and multi-energy flow supply data; A heat load prediction model is constructed, and a baseline heat load demand curve is obtained based on the multidimensional information data and multi-energy flow supply data using the heat load prediction model. Using the aforementioned baseline heat load demand curve as a constraint, an optimization model for the heat release plan of the thermal storage subsystem is constructed to obtain the optimal heat release power plan for each thermal storage unit. Construct a multi-source collaborative optimization model to obtain the optimal spatiotemporal matching scheme between multi-grade waste heat recovery, hierarchical storage and on-demand supply; The system assesses real-time heating capacity and switches the operating mode of the energy storage system based on process heat demand forecast data.

[0014] In another aspect, the present invention also provides a computer-readable storage medium storing instructions for causing a processor to execute a hybrid energy storage system for a continuous production industrial plant as described above.

[0015] The beneficial effects of this invention are: The embodiments of this invention construct a three-layer architecture consisting of an information perception and supply layer, a hybrid energy storage and conversion layer, and an optimization decision and control layer. Based on MES production scheduling plans, weather forecasts, and multi-grade waste heat data, a production plan-driven thermal energy storage base load scheduling method is formed, realizing the essential integration of energy scheduling and production processes. This enables energy supply to shift from passive energy supply to active matching of processes, significantly improving the planning and accuracy of energy use and reducing waste at its source.

[0016] The embodiments of this invention construct a hybrid energy storage architecture that includes a multi-grade thermal storage subsystem and an energy conversion unit, and propose a multi-source collaborative optimization model based on heat source temperature, flow rate and thermal grade parameters. This enables efficient cascade utilization of multi-grade waste heat, recovers waste heat at different temperatures in stages and accurately matches process requirements, effectively improving the overall waste heat recovery rate of the plant.

[0017] The present invention addresses the technical challenge of short battery lifespan caused by deep charging and discharging in industrial scenarios by introducing a virtual lifespan loss cost proportional to the absolute value of battery charging and discharging power into the system economic optimization model and using this model for scheduling decisions. It also makes the battery tend to charge and discharge smoothly and with low power from the algorithmic source, thus extending its cycle life.

[0018] The embodiments of this invention define a heat gap criterion based on real-time monitoring and design a dual-use switching logic that includes emergency discharge of electrochemical energy storage, dynamic adjustment of thermal storage priority, and rapid replenishment of energy storage gap. This constructs a thermal energy resilience guarantee system that fits the actual production situation. The emergency response directly targets threats to production continuity, ensuring more precise and effective protection.

[0019] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0020] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a diagram illustrating the architecture of a hybrid energy storage system for a continuous production industrial plant according to one embodiment of the present invention. Figure 2 A flowchart illustrating heat load prediction by a heat load prediction module according to an embodiment of the present invention; Figure 3 A flowchart illustrating the multi-grade waste heat recovery and cascade utilization scheduling process according to an embodiment of the present invention; Figure 4 This is a flowchart of a method for switching between normal and emergency use based on a heat gap criterion according to an embodiment of the present invention. Figure 5 This is a flowchart of a hybrid energy storage scheduling method for a continuous production industrial plant according to an embodiment of the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0022] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0023] like Figure 1 The diagram shows a hybrid energy storage system architecture for a continuous production industrial plant according to one embodiment of the present invention. Figure 1 The system may include an information sensing and supply layer, a hybrid energy storage and conversion layer, and an optimization decision and control layer connected in sequence.

[0024] The information sensing and supply layer is used to acquire multi-dimensional information data and multi-energy flow supply data. In this embodiment, the information sensing and supply layer is divided into data sources and energy sources. Data sources are used to acquire multi-dimensional information data, and energy sources are used to acquire multi-energy flow supply data. Data sources include factory manufacturing execution systems (MES), weather forecasting systems, equipment performance parameters, sensors, etc.; energy sources include solar photovoltaic power generation systems, solar thermal systems, municipal power grids, and multi-grade process waste heat from the factory (such as high-temperature flue gas, medium-temperature equipment cooling water, and low-temperature air conditioning return water).

[0025] The hybrid energy storage and conversion layer is used to realize multi-grade thermal energy storage, electrochemical energy storage, and thermoelectric coupling conversion. In this embodiment, the hybrid energy storage and conversion layer is divided into a multi-grade thermal energy storage subsystem, an electrochemical energy storage subsystem, and an energy conversion unit. The multi-grade thermal energy storage subsystem includes multiple thermal energy storage units (such as phase change hot water tanks), each of which is independently equipped with a circulation pump, temperature sensor, and control valve. The electrochemical energy storage subsystem is mainly a battery energy storage system, whose converter (PCS) has rapid power regulation capability. The energy conversion unit mainly consists of an electrically driven heat pump (such as a water source heat pump, an air-cooled heat pump, etc.) as a bridge connecting electrical energy and thermal energy of various grades.

[0026] The optimization decision and control layer is used to execute scheduling optimization algorithms. In this embodiment, the optimization decision and control layer includes a heat load prediction module, a multi-source collaborative optimization model module, and an operating mode switching module. Specifically, the heat load prediction module predicts the optimal heat release power plan for each thermal storage unit based on data obtained from the information sensing and supply layer; the multi-source collaborative optimization model module solves for the optimal spatiotemporal matching scheme between multi-grade waste heat recovery, tiered storage, and on-demand supply; and the operating mode switching module evaluates the real-time heating capacity and switches the operating mode of the energy storage system based on process heat demand prediction data.

[0027] In this embodiment, the heat load prediction module includes a heat load prediction model and a heat release plan optimization model for the thermal storage subsystem. The process of heat load prediction by the heat load prediction module is as follows: Figure 2 As shown, the MES production scheduling plan is used as the core input. A baseline heat load demand curve is generated through the heat load prediction model. This curve is then used as a rigid constraint to optimize the heat release plan of the thermal storage subsystem, thereby achieving proactive matching between energy supply and production processes.

[0028] Specifically, in this example, the production schedule for the next 24 hours is first obtained from the factory's Manufacturing Execution System (MES), including the start and stop times of each process unit, capacity load, and corresponding heat demand parameters. Combined with ambient temperature and solar irradiance data provided by the weather forecasting system, a heat load prediction model is established. The heat load prediction model is a hybrid model integrating physical mechanisms and data-driven approaches. Its inputs include at least: production plan characteristics (production line start / stop status, product type, planned output, process setting parameters), meteorological characteristics (ambient temperature, humidity, solar irradiance), and historical heat load data. The model output is the baseline heat load prediction values ​​and their confidence intervals for each future time period.

[0029] Production planning features, including production line start-up and shutdown time matrices, can be obtained from the MES (Manufacturing Execution System). Product type coding Planned production Process setting temperature and traffic Environmental characteristics, including the temperature over the next 24 hours, are obtained from meteorological systems. ,humidity Solar irradiance Historical heat loads for the corresponding time period are extracted from the historical database. Actual production data and meteorological data.

[0030] The hybrid model, which integrates physical mechanisms and data-driven approaches, is structured into static terms. With dynamic items : ; The static terms are based on the physical model of the production plan: ; in, , The production line heat consumption coefficient is calibrated using historical data; This represents the total number of production lines within the factory that participate in the heat load calculation. For production line During the period The running status indicator variable, whose value comes directly from the MES production scheduling plan; For production line During the period Yield; For production line The process setting temperature, For time period Ambient temperature; For production line The characteristic coefficient of heat loss at a given temperature.

[0031] Dynamic items It is a correction term based on LSTM, and its input feature vector is: It outputs a prediction of dynamic heat load deviation.

[0032] Random items To model Gaussian process regression (GPR) and quantify prediction uncertainty, an LSTM and GPR model were trained using historical data, and Bayesian optimization was employed for parameter tuning. Validation metrics were RMSE and MAPE, with MAPE required to be <5%.

[0033] Then, using the baseline heat load demand curve as a rigid constraint, an optimization model for the heat release plan of the thermal storage subsystem is established. The model aims to minimize the operating cost of the thermal storage system, and the objective function considers the power consumption of the circulating pump and the heat loss cost of the thermal storage unit. The constraints include: heat balance constraints (the total heat release of the thermal storage unit must meet the baseline heat load demand), dynamic energy balance constraints of the thermal storage unit, capacity and power limits for charging and releasing heat of the thermal storage unit, temperature limits of the thermal storage unit, and process temperature matching requirements.

[0034] Based on prediction To constrain the process, we establish an optimization problem for the thermal storage and release plan, with the objective function: ; in, This is the unit power consumption cost coefficient for the circulating pump. In order to be in Time period The operating power of the circulation pump of each thermal storage unit The cost coefficient per unit heat loss. In order to be in Time period Heat loss of each thermal storage unit.

[0035] Constraints are divided into: Thermal equilibrium constraint: ,in, For time period within, no. The actual effective heat released by each thermal storage unit.

[0036] Dynamic model of thermal storage unit (considering temperature stratification): ,in, For time period within, no. Total heat storage capacity of each heat storage unit; For time period within, no. The thermal power supplied to each thermal storage unit; For time period within, no. The temperature of each thermal storage unit; For time period Ambient temperature; The thermal resistance of the thermal storage unit, For the first The surface area of ​​each thermal storage unit.

[0037] Temperature and capacity limitations: ,in, For the first The minimum permissible operating temperature for each thermal storage unit. For the first The maximum allowable operating temperature of each thermal storage unit; For the first The minimum allowable energy storage capacity of a thermal storage unit. For the first The maximum energy storage capacity allowed per thermal storage unit.

[0038] Process temperature matching constraints: ,in, For time period within, no. The heat release temperature of each thermal storage unit; For time period The minimum heating temperature required for the process flow.

[0039] The optimization model employs a rolling solution using the Model Predictive Control (MPC) method to obtain the optimal heat release power plan for each thermal storage unit. During the solution process, the prediction model parameters are updated periodically (e.g., every 15 minutes) and re-optimized based on the deviation between the actual heat load and the predicted value to adapt to production fluctuations and weather changes.

[0040] Finally, the optimized heat release plan is distributed to the actuators (circulation pumps, control valves) of each thermal storage unit, achieving proactive matching between the thermal storage base load and the production plan. Through this method, energy supply shifts from a passive response to actively following the production rhythm, significantly improving the planning and accuracy of energy use.

[0041] In this embodiment, the multi-source collaborative optimization model module is used to construct a multi-temporal-scale collaborative optimization model (multi-source collaborative optimization model), which integrates discretely distributed waste heat recovery devices, multi-level thermal storage units with temperature stratification characteristics, and heat load points with specific temporal and spatial requirements into a unified optimization framework for decision-making.

[0042] Specifically, in this example, the scheduling process for multi-grade waste heat recovery and cascade utilization is as follows: Figure 3 As shown, firstly, a digital map of the entire plant's waste heat resources and heat load is established. Parametric models are then created for each waste heat recovery point within the plant (such as high-temperature flue gas, medium-temperature equipment cooling water, and low-temperature air conditioning return water), recording its heat source temperature grade, maximum recoverable power, geographical location, and recoverable time period. Simultaneously, process heat load points are modeled, clarifying their required temperature, required power curves, required time windows, and geographical locations. The specific parametric modeling is as follows: Waste heat recovery device: for the first There are [number] recycling points, with a maximum recycling power of [number]. The temperature of the heat source is Define its switch state binary variable. and recovery power continuous variable Introducing start-up and shutdown costs This is to punish frequent start-stop operations.

[0043] Multi-grade thermal storage units: classified according to temperature range as follows The thermal state of the first-level thermal storage unit is as follows: satisfy: ; in, This is the set of recovery points that can release heat to this unit. For heat charging efficiency, For heat release power, This is due to heat loss.

[0044] Heat demand node: the first Each process requirement point, in time period Required temperature is The power is the amount of heat. This heat must be generated by a source at a temperature not lower than [a certain value]. The thermal storage unit or recovery device supplies the heat directly (grade matching constraint).

[0045] Then, a multi-source collaborative optimization model is constructed with the objective of maximizing the overall comprehensive benefit of the system. Its objective function (maximizing the overall system benefit) is: ; in, The value of thermal energy (which can be converted into savings in gas or steam costs). In order to be in Time period The actual heat demand satisfied at each heat load node. In order to be in Time period Start-up and shutdown status variables of a waste heat recovery device For the first The cost of starting and stopping a single waste heat recovery device In order to be in Time period The recovery power of each waste heat recovery device In order to be in Time period Heat loss of each thermal storage unit For the first Unit power operating cost of a waste heat recovery device The cost coefficient per unit heat loss. This represents the total number of time periods within the scheduling cycle.

[0046] This model enables spatiotemporal matching of heat grade, internalization of start-up and shutdown costs, and multi-level thermal storage mediation. Specifically: Spatiotemporal matching of heat grade ensures that, during any scheduling period, the heat supplied to any heat load point must have a temperature grade no lower than the minimum requirement of that load point, and the heat source (direct recovery or thermal storage release) must be accessible in both time and space. Internalization of start-up and shutdown costs explicitly includes the efficiency loss costs incurred by each waste heat recovery device due to state switching (start-up and shutdown) in the model's objective function, thus avoiding unrealistic frequent start-up and shutdown commands in the optimization results. Multi-level thermal storage mediation uses thermal storage units of different temperature levels (high, medium, and low) as key spatiotemporal buffers and grade conversion mediators. Optimization decisions include not only when and how much heat to recover, but also which grade of heat to store in which level of thermal storage unit and when to release heat from which level of thermal storage unit to meet which demand, thereby shifting waste heat in the time dimension and achieving targeted utilization in the grade dimension.

[0047] The decision variables of the multi-source collaborative optimization model include the start-up and shutdown status of each waste heat recovery device, the recovered power and its allocation (direct supply to the load or storage in a thermal storage unit of a specified grade) during each future scheduling period, as well as the charging and discharging heat power of each thermal storage unit. The core constraints include: Grade matching and heat flow balance constraints: ; in, Indicates in Time period The thermal storage unit sends to the first The effective heat supply released by each heat load node For the first The current available temperature rating of each thermal storage unit. For the first The minimum supply temperature required by each heat load node. For the first The supply temperature level of a direct heat source, The heat transport efficiency coefficient of a direct heat source. In order to be in Time period The first direct heat source to the first The original heat power delivered by each heat load node Indicates in Time period The total heat demand of each heat load node. This constraint ensures that the supplied heat meets the demand in terms of both temperature and quantity.

[0048] Power recovery allocation and logic constraints: , Ensure that the recovered power does not exceed the upper limit, and that the sum of the allocated destinations (direct heating or storage) equals the total recovered amount.

[0049] Dynamics and capacity constraints of thermal storage units: , .

[0050] Spatiotemporal transmission loss: For collection points and demand points that are far apart, a pipeline loss coefficient is introduced. The portion of direct heating is reduced.

[0051] Next, a rolling optimization framework is used to solve the model. Taking a typical day as an example, the equipment start-up and power allocation plan for the next 24 hours is solved at hourly intervals, and a rolling correction is performed every 15 minutes based on actual operating deviations (such as heat source fluctuations and demand fine-tuning), outputting executable instructions.

[0052] Finally, optimization commands are sent to the actuators of each recovery device, heat storage unit, and pipeline valve, enabling dynamic optimization and efficient utilization of waste heat resources across the plant in terms of time, space, and temperature grade. This method maximizes the recovery rate of low-grade waste heat, reduces the downgrading of high-grade waste heat, and improves the overall energy utilization efficiency of the plant.

[0053] In this implementation, a thermally resilient emergency support system is constructed for continuous production processes. The working mode switching module adopts a dual-use switching method based on heat gap criteria. Normal mode: With the goal of economic optimization, a forecast-based scheduling plan is executed. Emergency mode: With the highest priority goal of ensuring the continuity of the core process heat load, all available resources are mobilized to the maximum extent for heat replenishment. The mode switching is automatically triggered by a heat gap criterion based on real-time monitoring and ultra-short-term forecasting. The calculation of the heat gap is based on the real-time comparison of two key parameters: (1) Real-time maximum reliable heating capacity: The system comprehensively evaluates the maximum total heat power that all heat sources (including the real-time heat release capacity of each grade of heat storage unit and the maximum heating capacity of electric heat pumps supported by available electricity) can provide in a short period of time in the future. (2) Ultra-short-term heat load demand: The system makes rolling predictions of the process heat load demand in the same short period of time in the future based on real-time sensor data and process operating status. The heat gap is the difference between the above-mentioned predicted demand and the maximum heating capacity. When the system detects that the heat deficit exceeds the preset safety threshold, it will immediately switch from normal mode to emergency mode.

[0054] In this example, the flowchart of the emergency / restaurant switching method based on the heat gap criterion is as follows: Figure 4As shown, the operating mode switching module includes a real-time monitoring and prediction module and a threshold judgment module. The real-time monitoring and prediction module is used to assess the real-time heating capacity and calculate the heat gap. Specifically, the real-time heating capacity needs to be assessed first. The system calculates the maximum reliable heating capacity for a short future period (e.g., 30-60 minutes) in real time (e.g., every 5 minutes). ; in, For the first Each thermal storage unit in the future The maximum sustainable heat release capacity within a given time period depends on its current heat storage capacity, temperature, and the limits of the heat release equipment. This is the rated electrical power of the heat pump. For the future The maximum reliable electrical power that can be supplied to the heat pump during a given period is derived from the minimum of the real-time output of photovoltaic power, the emergency reserve power of the battery, and the available grid capacity (considering contractual limits). This is the coefficient of performance (COP) of the heat pump.

[0055] Based on recent actual load data, we will make rolling forecasts of process heat demand in the near future. Calculate the heat deficit: .

[0056] The threshold determination module is used to determine the working model to be used in the current cycle based on the heat gap. In this example, the determination logic of the threshold determination module may include: In step S10, it is determined whether the heat deficit is greater than the warning threshold; In step S11, if the heat gap is determined to be greater than the warning threshold, the warning state is triggered, the equipment status is checked, and the emergency logic is prepared to be started. In step S12, it is determined whether the heat deficit is greater than the emergency trigger threshold; In step S13, if the heat gap is determined to be greater than the emergency trigger threshold, the energy storage system is controlled to switch from normal mode to emergency mode; in the emergency mode, battery protection mode switching, heat storage strategy reconstruction and gap filling calculation operations are performed. In step S14, it is determined whether the heat gap has been eliminated; In step S15, if it is determined that the heat gap has not been eliminated, the emergency mode is maintained. If the heat deficit is determined to have been eliminated, the energy storage system is switched to normal mode.

[0057] Steps S10 to S15 involve setting two levels of trigger thresholds: a warning threshold and a warning threshold. and emergency trigger threshold .when When this happens, the system enters an early warning state, checks the equipment status in advance, and prepares to activate emergency logic; when The system switches from normal mode to emergency mode.

[0058] In emergency mode, the scheduling objective shifts from optimal economic efficiency to optimal production continuity, and the following emergency collaborative scheduling logic is executed: 1. Emergency mobilization of power resources: Instruct the electrochemical energy storage system to immediately switch to the power supply guarantee mode, and provide stable power to key heat source equipment such as electric heat pumps at its maximum capacity or as needed to ensure their full operation.

[0059] 2. Prioritization of thermal storage resources: Dynamically adjust the heat release strategy of all thermal storage units, stop supplying heat to general loads, concentrate all thermal storage capacity, and prioritize the heat demand of core production links according to the preset priority order of process importance.

[0060] 3. Emergency and Recovery Linkage: While emergency dispatch is being executed, the system activates a heat gap filling calculation model. After the emergency event subsides, this model will automatically plan to charge the thermal storage system with the highest priority during a suitable subsequent period (such as when electricity prices are flat or photovoltaic output is sufficient), so that it can be restored to a safe thermal storage level capable of coping with the next event as soon as possible.

[0061] Finally, once the heat deficit is eliminated and the system is operating stably, it automatically switches back to normal mode. This method constructs a proactive thermal emergency response system deeply integrated with the production process, significantly improving the energy resilience and reliability of continuous production.

[0062] Furthermore, by quantifying the battery's lifespan degradation process into a real-time economic cost and embedding it into the scheduling optimization objective, proactive management of battery health is achieved from the decision-making source. Specifically, in this example, firstly, a lifespan degradation-economic cost mapping model for the battery energy storage system is established. This model can assess the instantaneous loss caused by the current scheduling action to the battery's cycle life based on the battery's real-time operating status (including but not limited to state of charge, charge / discharge power, and temperature), and quantify this physical loss into a corresponding virtual economic cost. The virtual cost is correlated with the real-time charge / discharge power.

[0063] Based on experimental data and a semi-empirical model, an instantaneous aging rate model characterized by a stress factor is established: ; in, In a charged state, For charging and discharging current (and power) (related) Battery temperature. Stress factor function. It typically increases significantly in the high / low SOC range and under high current conditions.

[0064] Transform physical aging into an economic cost that can be handled by an optimized model. Define the virtual lifetime depreciation cost rate. (RMB / kWh): ; in, This represents the initial total investment cost of the battery energy storage system. This refers to the rated number of cycles the battery performs under standard operating conditions. For the battery's rated capacity, Round-trip efficiency Let be the aging rate under standard operating conditions. Then, the lifespan loss cost caused by scheduling decisions during time period t is: ; based on It is a dynamic variable whose value changes with... , The algorithm adapts to real-time changes. When the battery is under high stress, the cost rate automatically increases, and the algorithm will automatically avoid high-power scheduling in this state.

[0065] Then, when formulating the overall system economic scheduling plan, the aforementioned virtual cost of battery life loss is included as a key cost item, and incorporated into the optimization objective function along with electricity purchase costs and equipment operation and maintenance costs for comprehensive minimization. This allows the optimization algorithm to automatically and intrinsically weigh the electricity price difference gained from battery charging and discharging against the long-term costs brought about by battery life loss when making economic decisions, thereby avoiding uneconomical and high-loss actions by the battery from the source.

[0066] Specifically, the optimization model of this method achieves battery protection through the following mechanism: State-adaptive cost: The virtual lifetime loss cost is not a fixed value, but is dynamically adjusted according to the real-time state of the battery (such as being in a high-charge state or a low-charge state). When the algorithm determines that the battery is in a high-stress state range that is prone to accelerated aging, it will automatically assign a higher virtual cost to the charging and discharging behavior during that period, thereby guiding the algorithm to reduce or avoid using the battery in that state.

[0067] Smooth power orientation: Since the cost of life loss is related to the absolute value of charge and discharge power, the optimization algorithm, in order to minimize the total cost, will naturally tend to plan a charge and discharge curve with smooth power changes and low peak values ​​for the battery, so as to avoid frequent high power surges.

[0068] Operating range optimization: The model can dynamically set its recommended state of charge operating range based on the battery's health status, guiding the battery to work in a comfortable range that is conducive to extending its lifespan.

[0069] The embedded co-optimization model for lifetime cost is to The modified objective function is embedded in a system-level optimization model that aims to minimize daily operating costs: ; in, For time-of-use electricity pricing, For the power purchase capacity, Other operation and maintenance costs.

[0070] The system power balance constraint is: ,in Photovoltaic power generation capacity, The power consumption of the heat pump is strongly coupled with the thermal storage scheduling. Basic electrical load; For other adjustable electrical loads.

[0071] Battery-specific constraints include power and SOC limits, as well as health-based soft SOC constraints. The power and SOC limits are as follows: The health-based SOC soft constraint is as follows: In the optimization model, the traditional fixed upper and lower limits of SOC are set as dynamically shrinkable soft constraints. For example, when the battery temperature... When the SOC is too high, the system automatically tightens the operating range to [value]. To avoid the high / low SOC stress zone at high temperatures, this constraint is achieved through a penalty function.

[0072] Finally, the optimization model, which includes a lifespan cost term, is solved to obtain the scheduling instructions for the battery energy storage system and other devices within the system (such as heat pumps). Through this method, the battery's role shifts from a traditional energy transporter to a system power regulator and flexibility provider, significantly extending the actual lifespan of the battery energy storage system while ensuring overall economic efficiency and improving the project's overall lifecycle economics.

[0073] On the other hand, the present invention also provides a hybrid energy storage scheduling method for a continuous production industrial plant. This method applies any of the hybrid energy storage systems described above for continuous production industrial plants, and may include, for example: Figure 5 The steps shown are described in this. Figure 5 In this context, the method may include: In step S20, multi-dimensional information data and multi-energy flow supply data are acquired; In step S21, a heat load prediction model is constructed, and the benchmark heat load demand curve is obtained based on multidimensional information data and multi-energy flow supply data using the heat load prediction model. In step S22, an optimization model for the heat release plan of the thermal storage subsystem is constructed using the baseline heat load demand curve as a constraint to obtain the optimal heat release power plan for each thermal storage unit. In step S23, a multi-source collaborative optimization model is constructed to obtain the optimal spatiotemporal matching scheme between multi-grade waste heat recovery, hierarchical storage and on-demand supply. In step S24, the real-time heating capacity is evaluated and the working mode of the energy storage system is switched according to the process heat demand forecast data.

[0074] In another aspect, the present invention also provides a computer-readable storage medium storing instructions for causing a processor to execute a hybrid energy storage system for a continuous production industrial plant as described above.

[0075] The beneficial effects of this invention are: The embodiments of this invention construct a three-layer architecture consisting of an information perception and supply layer, a hybrid energy storage and conversion layer, and an optimization decision and control layer. Based on MES production scheduling plans, weather forecasts, and multi-grade waste heat data, a production plan-driven thermal energy storage base load scheduling method is formed, realizing the essential integration of energy scheduling and production processes. This enables energy supply to shift from passive energy supply to active matching of processes, significantly improving the planning and accuracy of energy use and reducing waste at its source.

[0076] The embodiments of this invention construct a hybrid energy storage architecture that includes a multi-grade thermal storage subsystem and an energy conversion unit, and propose a multi-source collaborative optimization model based on heat source temperature, flow rate and thermal grade parameters. This enables efficient cascade utilization of multi-grade waste heat, recovers waste heat at different temperatures in stages and accurately matches process requirements, effectively improving the overall waste heat recovery rate of the plant.

[0077] The present invention addresses the technical challenge of short battery lifespan caused by deep charging and discharging in industrial scenarios by introducing a virtual lifespan loss cost proportional to the absolute value of battery charging and discharging power into the system economic optimization model and using this model for scheduling decisions. It also makes the battery tend to charge and discharge smoothly and with low power from the algorithmic source, thus extending its cycle life.

[0078] The embodiments of this invention define a heat gap criterion based on real-time monitoring and design a dual-use switching logic that includes emergency discharge of electrochemical energy storage, dynamic adjustment of thermal storage priority, and rapid replenishment of energy storage gap. This constructs a thermal energy resilience guarantee system that fits the actual production situation. The emergency response directly targets threats to production continuity, ensuring more precise and effective protection.

[0079] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0084] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0085] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0086] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0087] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A hybrid energy storage system for a continuous production industrial plant, characterized in that, The system comprises: an information sensing and supply layer, a hybrid energy storage and conversion layer, and an optimization decision-making and control layer, connected sequentially; wherein... The information sensing and supply layer is used to acquire multi-dimensional information data and multi-energy flow supply data; The hybrid energy storage and conversion layer is used to realize multi-grade thermal energy storage, electrochemical energy storage and thermoelectric coupling conversion; The optimization decision and control layer is used to execute the scheduling optimization algorithm; the optimization decision and control layer includes a heat load prediction module, a multi-source collaborative optimization model module, and a working mode switching module; the heat load prediction module is used to predict the optimal heat release power plan of each thermal storage unit based on the data obtained by the information perception and supply layer; the multi-source collaborative optimization model module is used to solve the optimal spatiotemporal matching scheme between multi-grade waste heat recovery, graded storage, and on-demand supply; the working mode switching module is used to evaluate the real-time heating capacity and switch the working mode of the energy storage system according to the process heat demand prediction data.

2. The system according to claim 1, characterized in that, The heat load prediction module includes a heat load prediction model and a heat release plan optimization model for the thermal storage subsystem, wherein, The heat load forecasting model is used to obtain the baseline heat load forecast values ​​and their confidence intervals for each future period, so as to obtain the baseline heat load demand curve. The heat load forecasting model includes static and dynamic terms. The static term is a physical model based on the production plan, and the dynamic term is a dynamic heat load deviation forecast based on the LSTM model. The thermal storage subsystem heat release plan optimization model is used to obtain the optimal heat release power plan for each thermal storage unit. The thermal storage subsystem heat release plan optimization model includes a cost minimization objective function and constraints. The constraints include thermal balance constraints, dynamic energy balance constraints of thermal storage units, capacity and charge / release power limits of thermal storage units, temperature limits of thermal storage units, and process temperature matching requirements.

3. The system according to claim 2, characterized in that, The cost minimization objective function is: ; in, This is the unit power consumption cost coefficient for the circulating pump. In order to be in Time period The operating power of the circulation pump of each thermal storage unit The cost coefficient per unit heat loss. In order to be in Time period Heat loss of each thermal storage unit.

4. The system according to claim 1, characterized in that, The multi-source collaborative optimization model module includes a multi-source collaborative optimization model with the objective of maximizing the overall comprehensive benefit of the system. The objective function of the multi-source collaborative optimization model is: ； in, For thermal energy value, In order to be in Time period The actual heat demand satisfied at each heat load node. In order to be in Time period Start-up and shutdown status variables of a waste heat recovery device For the first The cost of starting and stopping a single waste heat recovery device In order to be in Time period The recovery power of each waste heat recovery device In order to be in Time period Heat loss of each thermal storage unit For the first Unit power operating cost of a waste heat recovery device The cost coefficient per unit heat loss. This represents the total number of time periods within the scheduling cycle.

5. The system according to claim 4, characterized in that, The constraints of the multi-source collaborative optimization model include grade matching and heat flow balance constraints, recovery power allocation and logical constraints, thermal storage unit dynamic and capacity constraints, and spatiotemporal transmission losses; among which, The constraints of quality matching and heat flow balance are: ； in, Indicates in Time period The thermal storage unit sends to the first The effective heat supply released by each heat load node For the first The current available temperature rating of each thermal storage unit. For the first The minimum supply temperature required by each heat load node. For the first The supply temperature level of a direct heat source, The heat transport efficiency coefficient of a direct heat source. In order to be in Time period The first direct heat source to the first The original heat power delivered by each heat load node Indicates in Time period Total heat demand of each heat load node.

6. The system according to claim 1, characterized in that, The working mode switching module includes a real-time monitoring and prediction module and a threshold judgment module; the real-time monitoring and prediction module is used to evaluate the real-time heating capacity and calculate the heat gap; the threshold judgment module is used to determine the working model to be used in the current cycle based on the heat gap.

7. The system according to claim 6, characterized in that, The judgment logic of the threshold judgment module includes: Determine whether the heat deficit is greater than the warning threshold; If the heat deficit is determined to be greater than the warning threshold, a warning state is triggered, the equipment status is checked, and emergency logic is prepared to be started. Determine whether the heat deficit is greater than the emergency trigger threshold; If the heat gap is determined to be greater than the emergency trigger threshold, the energy storage system is controlled to switch from normal mode to emergency mode; in the emergency mode, battery protection mode switching, heat storage strategy reconstruction and gap filling calculation operations are performed. Determine if the heat deficit has been eliminated; If the heat deficit is not eliminated, the emergency mode is maintained. If the heat deficit is determined to have been eliminated, the energy storage system is switched to normal mode.

8. The system according to claim 7, characterized in that, The battery protection mode switching includes: A collaborative optimization model for lifetime cost is constructed, and the objective function of the collaborative optimization model for lifetime cost is: ； in, The constraints of the collaborative optimization model for life cost include system power balance constraints and battery self-constraints.

9. A hybrid energy storage scheduling method for a continuous production industrial plant, characterized in that, The hybrid energy storage scheduling method for the continuous production industrial plant applies the hybrid energy storage system of the continuous production industrial plant as described in any one of claims 1-8, and the hybrid energy storage scheduling method for the continuous production industrial plant includes: Acquire multi-dimensional information data and multi-energy flow supply data; A heat load prediction model is constructed, and a baseline heat load demand curve is obtained based on the multidimensional information data and multi-energy flow supply data using the heat load prediction model. Using the aforementioned baseline heat load demand curve as a constraint, an optimization model for the heat release plan of the thermal storage subsystem is constructed to obtain the optimal heat release power plan for each thermal storage unit. Construct a multi-source collaborative optimization model to obtain the optimal spatiotemporal matching scheme between multi-grade waste heat recovery, hierarchical storage and on-demand supply; The system assesses real-time heating capacity and switches the operating mode of the energy storage system based on process heat demand forecast data.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions for causing a processor to execute a hybrid energy storage system for a continuous production industrial plant as described in any one of claims 1-8.

Images