Dual-layer green electricity fluctuation buffer process and system based on phase change energy storage-multi-hearth coordination

Abstract

The application provides a double-layer green electricity fluctuation buffering process and system based on phase change energy storage-multiple-hearth coordination, S1, embedding a PCM heat storage layer in a coke oven carbonization chamber wall to form a three-layer composite structure of a refractory working layer, a PCM heat storage layer and a thermal insulation layer, and establishing a quantitative relationship between PCM heat storage and single-hearth supportable buffering time; S2, obtaining a power prediction curve, decomposing the power prediction curve into multiple intrinsic mode functions by using variational mode decomposition, and dividing the power prediction curve into fast components and slow components according to the period; S3, isothermally buffering the fast components through the endothermic melting or exothermic solidification of the phase change material; S4, power distribution for the slow components; S5, dynamically distributing power by comprehensively considering the thermal response state of the PCM heat storage layer in S3 and the power adjustment margin of each hearth in S4. The application is suitable for large-scale coking enterprises driven by green electricity.

Description

Technical Field

[0001] This invention relates to the field of coking engineering technology, and more particularly to a dual-layer green electricity fluctuation buffering process and system based on phase change energy storage and multi-furnace coordination. Background Technology

[0002] With the deepening of the construction of new power systems, the penetration rate of green electricity in the industrial sector continues to increase. As one of the most carbon-intensive links in the steel industry chain, the coking industry relies on green electricity to replace fossil fuels for coke oven heating as a crucial technological path to achieve deep decarbonization. However, there is a contradiction between the inherent volatility of green electricity and the stringent temperature stability requirements of the coking process. Photovoltaic power generation can experience output fluctuations of over ±20% within seconds to minutes due to cloud cover, while wind power can fluctuate by ±50% within minutes to hours due to gusts or changes in wind direction. In contrast, temperature fluctuations in the coke oven carbonization chamber exceeding ±10 degrees Celsius can lead to a significant decline in coke quality. Therefore, achieving precise and stable temperature control of the coke oven under fluctuating green electricity conditions is a core issue facing the promotion of electric coke oven technology.

[0003] To address the aforementioned issues, domestic and international research institutions and enterprises have conducted research and developed relevant technical solutions in areas such as the design of electrically heated coke ovens, heating control and power regulation, phase change thermal storage materials and high-temperature applications, and power fluctuation decomposition and multi-timescale coordination. Patent document CN114317011A discloses an electrically heated coke oven structure that uses electric energy to heat the coal dry distillation process. The electric heating elements are arranged in the combustion chamber, solving the carbon emission problem of traditional coke oven gas heating mode and providing a basic structural design for the industrial application of electric coke ovens. Patent document CN117511569A adds a supplementary electric heating structure at the top of the oven, improving the uniformity of coke quality by optimizing the heat field distribution in the top area. Patent document CN120464428A discloses an electrically heated semi-coke oven and dry distillation method for low-rank pulverized coal, extending electric heating furnace technology to low-rank coal conversion applications. Patent document CN202016974U discloses a dual-channel supplementary heating device for coke oven heads, which improves the problem of low oven head temperature by adding an independent heating channel at the oven head. Patent document CN222064421U discloses a coke oven heating system with waste gas recycling, which improves thermal efficiency through waste gas recycling. In the field of phase change thermal energy storage materials, the academic paper "Review on Thermal Energy Storage with Phase Change Materials and Applications" systematically reviews the research progress of phase change thermal energy storage technology, covering medium-low temperature and medium-high temperature phase change materials, pointing out the potential of high temperature phase change materials in industrial applications and the engineering challenges faced when expanding to high-temperature industrial scenarios above 500 degrees Celsius. Patent document CN108413796A discloses a temperature gradient energy storage heat exchanger, which uses temperature gradients to achieve layered thermal storage of phase change materials. Its application scenario is a solar thermal utilization system, and the operating temperature range of the phase change material is 200 to 400 degrees Celsius. Patent document CN105509528A provides a multi-stage gradient phase change cold storage device, which utilizes phase change materials with different melting points to achieve graded cold storage, but it is geared towards low-temperature cold storage scenarios. Regarding power fluctuation decomposition and control strategies, variational mode decomposition (VMD) algorithms have been widely used in power system power fluctuation analysis. The academic paper "Coordinated Control of Hybrid Energy Storage System Based on Variational Mode Decomposition" proposes using VMD to decompose photovoltaic power fluctuations into high-frequency and low-frequency components, which are handled by supercapacitors and batteries respectively, achieving coordinated power allocation across multiple time scales.

[0004] In summary, while the structural design of electrically heated coke ovens has reached a certain level, none of the patented solutions address the synergistic issue of stable temperature control and power fluctuation buffering under fluctuating green electricity conditions. Their heating elements directly respond to grid power changes, making it difficult to maintain a constant carbonization chamber temperature under fluctuating green electricity scenarios. Existing coke oven control technologies primarily focus on temperature uniformity and combustion efficiency optimization under traditional gas heating methods, with limited attention paid to the control of electrically heated coke ovens under multi-timescale power fluctuation conditions. Furthermore, their control methods prioritize constant temperature control and do not consider dynamic buffering strategies for heat source power fluctuations. While there are precedents for using zonal arrangement of phase change materials to improve thermal storage efficiency, these solutions are all geared towards solar thermal utilization or low-temperature cold storage scenarios. A solution that embeds high-temperature phase change materials into the coke oven walls and utilizes the latent heat of phase change to buffer rapid fluctuations in green electricity has not yet been reported. Current research on variational mode decomposition (VMD) power allocation primarily focuses on electrochemical energy storage systems. However, the thermal response characteristics of phase change thermal storage materials differ fundamentally from those of electrochemical energy storage. Existing technologies have not combined VMD power decomposition with the buffering characteristics of phase change thermal storage to establish multi-timescale buffering strategies suitable for the characteristics of thermal equipment. In summary, in the scenario of green electricity-driven coke oven heating, the technical challenges of achieving rapid buffering of power fluctuations at the second to minute level at a single furnace level, achieving power matching of slow fluctuations at the minute to hour level at the furnace group level, and constructing a time-scale separated, two-layer collaborative buffering architecture remain unresolved in existing intellectual property. Summary of the Invention

[0005] In view of this, the purpose of this invention is to propose a dual-layer green electricity fluctuation buffering process and system based on phase change energy storage and multi-furnace coordination. The technical problems to be solved are how to achieve rapid buffering of green electricity fluctuations at the second to minute level at the single furnace level, how to achieve power matching of slow fluctuations at the minute to hour level at the furnace group level, and how to construct a dual-layer collaborative buffering architecture with time scale separation.

[0006] The technical means employed in this invention are as follows: A dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination includes the following steps: S1. Embed a PCM heat storage layer in the furnace wall of the coke oven carbonization chamber to form a three-layer composite structure of refractory working layer, PCM heat storage layer and heat insulation layer, and establish a quantitative relationship between PCM heat storage and buffer time that a single furnace can support. S2. Obtain the power prediction curve through the green electricity power prediction unit, and use the variational mode decomposition algorithm to decompose the power prediction curve into multiple intrinsic mode functions, based on the thermal response time constant of the PCM thermal storage layer in S1. Determine the boundary period, and divide each eigenmode function into fast and slow components according to the period, wherein the period of the fast component should be less than [period missing]. The period of the slow component should be greater than or equal to And the duration is longer than in S1. The fluctuations are coordinated by the furnace group; among them, the thermal response time constant It is only used to complete the division of fast and slow components; after the division is completed, the buffer time Δtb that a single furnace can support, calculated by S1, is used as the execution criterion for whether to start the coordinated control of the furnace group.

[0007] S3, Based on the PCM heat storage established in S1 and buffer time As a design benchmark, based on the phase change thermodynamic characteristics of the PCM thermal storage layer, a thermal power balance equation for the PCM layer is established. The rapid component is isothermally buffered on a timescale of seconds to minutes by the endothermic melting or exothermic solidification of the phase change material, with the buffering time not exceeding S1. ; S4. For the slow component identified in S2, when the duration of the slow component exceeds the buffer time that the PCM thermal storage layer can support in S3, a macroscopic slow fluctuation power matching model for the furnace group is established based on the power adjustment margin difference of different working states of multiple furnaces. The power adjustment margin of each furnace is calculated according to the working state coefficient of each furnace and the thermal state coefficient of the PCM thermal storage layer. The power allocation of the slow component is carried out at the minute to hour level through state priority coordination. S5. Establish a dual-layer collaborative operation strategy and PCM thermal state-driven power allocation decision. Combine the PCM layer thermal state, remaining heat storage and remaining supportable buffer time output by S3, and the power adjustment margin of each furnace output by S4 to perform dynamic power allocation.

[0008] Furthermore, in S1, the PCM thermal storage layer adopts a zoned selection strategy. Based on the temperature distribution characteristics of the carbonization chamber along the height direction, the PCM thermal storage layer is divided into three temperature zones, specifically including: In the high-temperature zone, with a temperature range of 800–1000 ℃, a NaCl-Na2SO4 mixed salt system is selected, with NaCl and Na2SO4 mixed in molar fractions of 60%–80% and 20%–40%, respectively. For the medium temperature range of 600–800 ℃, Al-Si eutectic alloy is selected. In the low-temperature zone, with a temperature range of 400–600 ℃, a binary molten salt of NaNO3-KNO3 is selected.

[0009] Furthermore, the quantitative relationship between the PCM heat storage and buffer time includes the PCM heat storage and the buffer time that a single furnace can support; the buffer time that a single furnace can support refers to the longest duration under the current PCM heat storage conditions when a single furnace can continuously withstand fluctuations in green power by relying solely on the PCM heat storage layer to absorb or release heat to maintain the stable temperature of the carbonization chamber, without needing to activate the coordinated control of the furnace group.

[0010] The formula for calculating the heat storage capacity of PCM is:

[0011] In the formula: Store heat for PCM; r PCM density; For PCM layer volume; L The latent heat of PCM; Fill rate; Buffer time that a single furnace can support The calculation formula is:

[0012] In the formula: Rated power for a single furnace chamber; This represents the proportion of power fluctuation amplitude that needs to be buffered.

[0013] Furthermore, in S2, the green power prediction unit adopts an LSTM-XGBoost combined model. The input variables of the green power prediction unit include historical power data, weather forecast data, and power grid dispatch plan. The output of the green power prediction unit is a power prediction curve.

[0014] Furthermore, S3 specifically includes the following steps: Establish the thermal power balance equation for the PCM layer:

[0015] In the formula: The effective heat capacity of the carbonization chamber; Input power to the heating element; This is due to heat loss; The heat absorption / release power of the PCM phase change; The phase change heat power of PCM is calculated using the equivalent heat capacity method, as shown in the following formula:

[0016] In the formula: δ represents the specific heat capacity of the solid state; L represents the latent heat; δ is the Gaussian distribution function, with its peak value located at the phase transition temperature. ; When the green electricity power suddenly drops by ΔP, the PCM layer compensates through heat release during solidification, as shown in the following formula:

[0017] In the formula: The thermal response time constant of the PCM layer; PCM layer thermal response time constant Determined by the thermal resistance and thermal capacity of the PCM layer:

[0018] In the formula: Equivalent heat capacity; PCM layer thickness; It is the equivalent thermal conductivity.

[0019] Furthermore, in S4, the calculation formula for the power regulation margin of each furnace is as follows: (1) Formula for calculating the upward adjustment margin:

[0020] (2) Formula for calculating the downward adjustment margin:

[0021] In the formula: This refers to the working status coefficient; The thermal state coefficient of the PCM; and These represent the maximum heat power that the PCM layer can release and absorb, respectively.

[0022] Furthermore, in S5, the two-layer collaborative operation strategy is as follows: When the fluctuation range of green electricity power Δ P <5% or duration Δ t When the temperature is less than 2 minutes, it is completely absorbed by the phase change thermal storage layer; When 5%≤Δ P <25% and 2 min≤Δ t When the time is less than 10 minutes, the phase change thermal storage layer bears 60% to 80% of the fluctuation, and the remainder is absorbed by the furnace group in a coordinated manner. When Δ P ≥25% or Δ t When the time is ≥10 min, the phase change thermal storage layer undertakes the rapid component for the first 10 to 20 minutes, and the subsequent slow component is undertaken by the furnace group in coordination. When all furnace PCM layers are in a thermally depleted state, the system degenerates into a pure multi-furnace coordination mode.

[0023] Furthermore, in S5, the priority order of the dynamic power allocation is as follows: The first priority is the furnace during the heating period and the PCM melting rate is <0.7; the second priority is the furnace during the heat preservation period and the PCM melting rate is >0.3; the third priority is the furnace during the coal charging period and the coking period; and the fourth priority is the furnace during the constant temperature period.

[0024] Furthermore, the power allocation weight formula for the dynamic power allocation is:

[0025] In the formula: For the first i The weighting of each furnace chamber is as follows: w=4 for the furnace chamber during the heating period, w=2 for the furnace chamber during the heat preservation period, w=1 for the furnace chamber during the coal charging and coking periods, and w=0.2 for the furnace chamber during the constant temperature period.

[0026] This invention also provides a dual-layer green electricity fluctuation buffering system based on phase change energy storage and multi-furnace coordination, for use in any one of the above-described dual-layer green electricity fluctuation buffering processes based on phase change energy storage and multi-furnace coordination, comprising: The green power prediction unit receives historical power data, weather forecast data, and power grid dispatch plan data at its input end. After processing the received data, it outputs the power prediction curve for the future period to the two-layer coordination controller. The dual-layer coordination controller has a first input terminal connected to the output terminal of the green electricity power prediction unit to obtain the power prediction curve, a second input terminal connected to the temperature sensor arranged at the interface of the phase change thermal storage layer of each furnace wall to obtain the real-time thermal state signal of the phase change thermal storage layer, and a third input terminal connected to the operating condition monitoring module of each carbonization chamber to obtain the operating status signal of each furnace. The dual-layer coordination controller runs a prediction, dual-layer scheduling and response control algorithm to generate power distribution instructions for each carbonization chamber based on the power prediction curve, the thermal state signal of the phase change thermal storage layer and the operating status signal of each furnace, and outputs the power distribution instructions to the electrothermal power distribution unit through the output terminal, while simultaneously outputting charge and discharge control instructions to the external electrochemical energy storage unit. The electric heating power distribution unit has its power input terminal connected to the green electricity input bus and its control signal input terminal connected to the power distribution command output terminal of the dual-layer coordination controller. The electric heating power distribution unit has multiple independent power channels, and the output terminal of each power channel is connected to the electric heating element of the corresponding carbonization chamber. The electric heating power distribution unit distributes the electrical energy of the green electricity input bus to the electric heating element of each carbonization chamber according to the set power of each channel according to the received power distribution command. The multi-furnace coke oven body contains multiple carbonization chambers. The furnace wall of each carbonization chamber adopts a three-layer composite structure of refractory working layer, phase change heat storage layer and heat insulation layer. The phase change heat storage layer is divided into three temperature zones along the height of the carbonization chamber: high temperature zone, medium temperature zone and low temperature zone. The high temperature zone is filled with sodium chloride-sodium sulfate mixed salt system, the medium temperature zone is filled with aluminum-silicon eutectic alloy, and the low temperature zone is filled with sodium nitrate-potassium nitrate binary molten salt. The phase change heat storage layer isothermally buffers the power fluctuations output from the electric heating power distribution unit to the electric heating element through the melting heat absorption or solidification heat release of the phase change material. The external electrochemical energy storage unit has its control signal input terminal connected to the charge and discharge control command output terminal of the dual-layer coordination controller, and its power terminal connected to the green electricity input bus. The external electrochemical energy storage unit charges or discharges according to the received charge and discharge control commands to supplement and buffer the power shortage or power excess outside the total power adjustment range of the furnace group.

[0027] Compared with the prior art, the present invention has the following advantages: This invention embeds a phase change heat storage layer within the furnace wall of a coke oven's carbonization chamber, creating a three-layer composite structure comprising a refractory working layer, a phase change heat storage layer, and a thermal insulation layer. The phase change heat storage layer utilizes the latent heat of phase change to form an isothermal buffer platform near the phase change temperature. When the green electricity power experiences rapid fluctuations on the order of seconds to minutes, the phase change material responds instantly through melting and heat absorption or solidification and heat release, ensuring that the carbonization chamber temperature is unaffected by instantaneous changes in electrical power. This design integrates the heat storage function with the coke oven wall, fundamentally solving the power-temperature mismatch problem caused by the direct response of heating elements to grid power changes without altering the coke oven's operating principle.

[0028] This invention upgrades the traditional two-layer furnace wall structure (refractory and insulation) to a three-layer composite structure (refractory, heat storage, and insulation). The heat storage layer is placed between the refractory and insulation layers, utilizing the high-temperature heat transferred from the refractory layer while reducing heat loss through the insulation layer. This creates a gentle isothermal plateau within the phase change heat storage layer. The latent heat storage density of phase change materials is significantly higher than that of sensible heat storage density of refractory bricks, allowing for the storage of more heat within the same storage space. Furthermore, the integrated design of the phase change heat storage layer and the furnace wall avoids the multi-stage heat exchange required by an external independent heat storage system, reducing intermediate heat exchange losses.

[0029] This invention proposes for the first time a time-scale separated, dual-layer collaborative buffering architecture. Through variational mode decomposition, green power fluctuations are periodically decomposed into fast and slow components, which are handled by a phase change thermal storage layer and a furnace group coordination layer, respectively. The phase change thermal storage layer, with its rapid thermal response characteristics, handles rapid fluctuations ranging from seconds to minutes, while the multi-furnace coordination layer utilizes the differences in power regulation margins between different furnace operating states to handle slow fluctuations ranging from minutes to hours. The two layers do not interfere with each other and complement each other, achieving time-scale separation and buffering of fluctuating energy, reducing capacity dependence on external electrochemical energy storage units.

[0030] The proposed power allocation strategy driven by the thermal state of the phase change thermal storage layer uses the melting rate of the phase change material as the key input variable for power allocation decisions. When the melting rate is in the middle range, the furnace has bidirectional buffering capability and prioritizes power allocation; when the melting rate is close to the limit, the furnace withdraws from or reduces power allocation. This strategy achieves a dynamic balance where furnaces with buffering capabilities withstand fluctuations, while furnaces without buffering capabilities maintain stable operation. It can predict the attenuation trend of the buffering capacity of each furnace and take preventive measures in advance, avoiding blind allocation that leads to overheating or undercooling of some furnaces, and improving the operational stability of the entire furnace group under fluctuating green electricity conditions.

[0031] This invention employs a zoned selection strategy, dividing the phase change heat storage layer into three temperature zones along the height of the carbonization chamber based on temperature distribution characteristics. The high-temperature zone uses a sodium chloride-sodium sulfate mixed salt system, the medium-temperature zone uses an aluminum-silicon eutectic alloy, and the low-temperature zone uses a sodium nitrate-potassium nitrate binary molten salt. The melting point gradients of the three phase change materials match the temperature gradient along the height of the carbonization chamber, achieving full coverage of phase change heat storage within the coke oven's operating temperature range. This ensures that the phase change materials in each temperature zone can fully utilize their latent heat of phase change within their respective operating temperature ranges. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a diagram of the overall system architecture of the present invention.

[0034] Figure 2 This is a schematic diagram of the composite furnace wall structure of the single carbonization chamber of the present invention.

[0035] Figure 3 This is a schematic diagram of the multi-furnace composite wall structure of the present invention.

[0036] Figure 4 This is a flowchart illustrating the dual-layer collaborative operation strategy and power allocation of the present invention. Detailed Implementation

[0037] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0038] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0039] This invention provides a dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination, comprising the following steps: S1. A PCM (high-temperature phase change) heat storage layer is embedded in the furnace wall of the coke oven carbonization chamber, forming a three-layer composite structure of a refractory working layer, a PCM heat storage layer, and a heat insulation layer. A quantitative relationship is established between the PCM heat storage capacity and the buffer time that a single furnace can support. The buffer time that a single furnace can support refers to the longest duration under the current heat storage conditions during which the PCM heat storage layer can maintain a stable furnace temperature solely through phase change heat absorption or release, without requiring the activation of furnace group coordination to continuously buffer fluctuations in green electricity power.

[0040] Specifically, the PCM heat storage layer is placed between the refractory working layer and the thermal insulation layer. Placing the PCM layer between the refractory and insulation layers allows the temperature gradient within the PCM layer to form an isothermal plateau with a gentle temperature change near the phase change temperature, which is beneficial for the full phase change of the PCM. The refractory working layer uses high-alumina refractory bricks with a working temperature range of 1000–1150℃ and a room-temperature thermal conductivity of approximately 1.5 W / (m·K), undertaking the main functions of temperature drop and high-temperature protection. The PCM heat storage layer utilizes the latent heat of phase change to absorb or release heat, buffering power fluctuations. The thermal insulation layer uses ceramic fiber felt and lightweight insulating bricks to reduce heat loss. This three-layer composite structure integrates the heat storage function into the furnace wall structure without increasing the overall thickness of the furnace wall.

[0041] The PCM thermal storage layer employs a zoned selection strategy along the height of the carbonization chamber. Based on the temperature distribution characteristics along the height of the carbonization chamber, the PCM thermal storage layer is divided into three temperature zones. The high-temperature zone, ranging from 800 to 1000℃, uses a NaCl-Na2SO4 mixed salt system, with NaCl and Na2SO4 mixed in molar fractions of 60%–80% and 20%–40%, respectively. This mixed salt system uses inexpensive and readily available raw materials and exhibits good chemical compatibility with refractory materials. The medium-temperature zone, ranging from 600 to 800℃, uses an Al-12Si eutectic alloy, a mature industrial casting aluminum alloy material. The low-temperature zone, ranging from 400 to 600℃, uses a NaNO3-KNO3 binary molten salt with a mass ratio of 60:40. This material has mature commercial applications in solar thermal power generation, low technical risk, and an upper operating temperature limit of approximately 600℃.

[0042] PCM encapsulation and fixation methods employ either metal microcapsule encapsulation or honeycomb ceramic carrier adsorption. The functions of the honeycomb ceramic carrier include: providing a large number of uniformly distributed PCM storage spaces; enhancing the effective thermal conductivity of the PCM layer; preventing leakage and flow of liquid PCM; and providing mechanical support to resist volumetric stress during thermal cycling.

[0043] The quantitative relationship between PCM heat storage and buffer time includes the PCM heat storage and the buffer time that a single furnace can support. The formula for calculating PCM heat storage is:

[0044] In the formula: For PCM heat storage, kWh; r For PCM density, [kg / m³] 3 ]; PCM layer volume [m 3 ]; L The latent heat of PCM is [kJ / kg]. This represents the fill rate.

[0045] Buffer time that a single furnace can support The calculation formula is:

[0046] In the formula: The rated power for a single furnace is [kW]. This represents the proportion of power fluctuation amplitude that needs to be buffered.

[0047] S2. Obtain the power prediction curve through the green power prediction unit, and use the variational mode decomposition algorithm to decompose the power prediction curve into multiple intrinsic mode functions. Then, divide each intrinsic mode function into fast component and slow component according to the period. The period of the fast component is less than the set boundary value, and the period of the slow component is greater than the set boundary value.

[0048] The green power prediction unit employs an LSTM-XGBoost combined model, with input variables including historical power data, weather forecast data, and grid dispatch plans. Specifically, the historical power data is data from the previous 24 hours at 15-minute resolution; the weather forecast data includes solar irradiance, wind speed, temperature, and cloud cover; and the grid dispatch plan is the day-ahead output curve. The output of the green power prediction unit is a power prediction curve for the next 4 hours, with a prediction accuracy of nRMSE < 15%.

[0049] The power prediction curve is decomposed into K intrinsic mode functions (IMFs) using a variational mode decomposition algorithm, mathematically expressed as:

[0050] In the formula: For the first k One intrinsic mode function; K To determine the number of decomposition layers, typically K =3~5.

[0051] For the thermal response time constant of the PCM thermal storage layer with a period less than S1 The fluctuation component, 10 to 20 times larger than normal, mainly originates from rapid meteorological changes such as cloud cover and gusts. This component is borne by the PCM layer, utilizing the isothermal buffering characteristics of the PCM's latent heat of phase change to achieve temperature stability on a timescale of seconds to minutes. For periods longer than S1, the buffering time supported by a single furnace is [not specified]. The fluctuation component mainly originates from weather system evolution and power grid dispatch adjustments. This component is handled by the furnace group coordination, utilizing the power regulation margin differences among multiple furnaces under different operating conditions to achieve power balance on a time scale of minutes to hours. For fluctuation components with periods between the two thresholds mentioned above, they are dynamically allocated to the PCM layer or the furnace group coordination layer based on their fluctuation amplitude and duration. The boundary period between the two layers is calculated in S1. Determined, value is The time should be 10 to 20 times (approximately 6 to 13 minutes), preferably 10 minutes; this value can be adjusted within the range of 5 to 20 minutes according to actual working conditions.

[0052] Furthermore, the single furnace's buffer time, calculated using S1, serves as the basis for switching between the two-layer control. When the duration of the power fluctuation corresponding to the rapid component does not exceed the single furnace's buffer time, only S3 is executed, with the PCM thermal storage layer handling the buffering. When the duration of the power fluctuation exceeds the single furnace's buffer time, after executing the S3 rapid response, S4 is initiated, with the furnace group continuing to bear the subsequent continuous power fluctuations.

[0053] S3. Based on the phase change thermodynamic characteristics of the PCM thermal storage layer, a thermal power balance equation for the PCM layer is established, and the rapid components are buffered at the second to minute level by the melting heat absorption or solidification heat release of the phase change material.

[0054] Establish the thermal power balance equation for the PCM layer:

[0055] In the formula: The effective heat capacity of the carbonization chamber is [J / K]. The input power for the heating element is [W]. For heat dissipation loss, [W]; The heat absorption / release power of the PCM phase change is [W].

[0056] The phase change heat power of PCM is calculated using the equivalent heat capacity method, as shown in the following formula:

[0057] In the formula: For solid-state specific heat capacity, [J / (kg)] K); L is latent heat, [J / kg]; δ is a Gaussian distribution function, with its peak value located at the phase transition temperature. .

[0058] When the green electricity power suddenly drops by ΔP, the PCM layer compensates through heat release during solidification, as shown in the following formula:

[0059] In the formula: Let be the thermal response time constant of the PCM layer, [s].

[0060] PCM layer thermal response time constant Determined by the thermal resistance and thermal capacity of the PCM layer:

[0061] In the formula: For equivalent heat capacity, [J / (kg)] K)]; The thickness of the PCM layer is [m]. The equivalent thermal conductivity [W / (m K).

[0062] The design specifications for the PCM micro-fast buffer are that, in the event of a sudden 25% drop in green power, the PCM layer can buffer temperature fluctuations to within ±3℃ within 1 minute; the maximum continuous buffering time of the PCM layer is approximately 28–53 minutes.

[0063] The quantitative relationship between PCM heat storage and buffer time is used in S3 to verify the design specifications of the PCM layer, namely, the maximum continuous buffer time should cover the expected longest fluctuation duration. This is calculated using S1. In S3, the buffer capacity is used as an evaluation benchmark to determine whether the design of the PCM layer meets the process requirements. After S3 is completed, the thermal state of the PCM layer, the remaining heat storage, and the remaining support buffer time are output as input parameters for dynamic power allocation in S5.

[0064] When the duration of the power fluctuation corresponding to the rapid component obtained from S2 decomposition does not exceed the single furnace support buffer time calculated by S1, the PCM thermal storage layer rapid buffer control is executed; based on the quantitative relationship between the PCM heat storage and the single furnace support buffer time established by S1, the PCM heat storage is used as the design benchmark, and the phase change heat absorption and release capacity of the PCM thermal storage layer is dynamically adjusted according to the real-time PCM heat storage, so that the PCM thermal storage layer can quickly absorb or release heat to buffer the green electricity power fluctuation; the duration of the power fluctuation corresponding to the rapid component does not exceed the single furnace support buffer time calculated by S1; after S3 is completed, the thermal state of the PCM thermal storage layer, the remaining heat storage, and the remaining single furnace support buffer time are output as input parameters for dynamic power allocation in S5.

[0065] S4. For the slow component identified in S2, when the duration of the slow component exceeds the buffer time that the PCM thermal storage layer can support in S3, a macroscopic slow fluctuation power matching model for the furnace group is established. The power adjustment margin of each furnace is calculated based on the working state coefficient of each furnace and the thermal state coefficient of the PCM thermal storage layer. The power allocation of the slow component is carried out on a minute to hour level through state priority coordination.

[0066] The working states of each carbonization chamber are divided into five types: coal charging period (S1), heating period (S2), constant temperature period (S3), heat preservation period (S4), and coking period (S5). Under different working states, the temperature control target and thermal inertia characteristics of the furnace are significantly different, and therefore the power regulation capability is also different.

[0067] The formulas for calculating the power regulation margin of each furnace are as follows:

[0068]

[0069] In the formula: The working state coefficients are S2=0.3, S3=0.05, and S4=0.2. The thermal state coefficient of PCM is given when the melting rate is between 0.3 and 0.7. =1.0, otherwise =0); and These represent the maximum heat power (kW) that the PCM layer can release and absorb, respectively.

[0070] The total power adjustment range of the furnace group is the sum of the adjustment margins of each furnace:

[0071] In the formula: The total power adjustment range for the furnace group is [kW]. For the regulation margin of a single furnace, [kW].

[0072] When the duration of the power fluctuation obtained from S2 exceeds the buffer time that a single furnace can support calculated in S1, coordinated control of the furnace group is executed. Power is allocated based on the working state of multiple furnaces and the thermal state of the PCM thermal storage layer. The power adjustment margin of each furnace is calculated according to the working state coefficient of each furnace and the thermal state of the PCM thermal storage layer output in S3. Coordinated control of the furnace group is performed according to the power adjustment margin so that each furnace can share the continuous power fluctuation. After S4 is completed, the power adjustment margin and power allocation results of each furnace are output as input parameters for dynamic power allocation in S5.

[0073] S5. Establish a dual-layer collaborative operation strategy and PCM thermal state-driven power allocation decision. Based on the PCM layer thermal state, remaining heat storage and remaining supportable buffer time output by S3, and the power adjustment margin of each furnace calculated by S4, dynamic power allocation is carried out in combination with the furnace working state.

[0074] The two-layer collaborative operation strategy is as follows: When the fluctuation range of green electricity power Δ P <5% or duration Δ t Within 2 minutes, the thermal energy is completely absorbed by the phase change storage layer, and the furnace group coordination layer remains inactive. The PCM layer utilizes... With a rapid response time of approximately 38 seconds, it can completely absorb fluctuations within 1 minute.

[0075] When 5%≤Δ P <25% and 2min≤Δ tWithin 10 minutes, the phase change thermal storage layer bears 60%–80% of the fluctuation, while the remaining 20%–40% is absorbed by the furnace group in coordination. The PCM layer responds quickly first, absorbing most of the power deficit or surplus within 1–2 minutes; the furnace group coordination layer starts power redistribution at the same time as the PCM layer responds, gradually taking over the portion that the PCM layer cannot continuously bear within 5–10 minutes.

[0076] When Δ P ≥25% or Δ t When the time is ≥10 minutes, the phase change thermal storage layer undertakes the initial 10 to 20 minutes of rapid heat storage, and the subsequent slow heat storage is undertaken by the furnace group in coordination. After the power is restored, the PCM layer is given priority to replenish power to restore the 50% melting rate reserve state.

[0077] When all furnace PCM layers are in a thermally depleted state, i.e. When the temperature is approximately 0, the system degenerates into a pure multi-furnace coordinated mode. At this point, the temperature fluctuation will increase to around ±8℃, but it is still better than the ±15~20℃ range without any buffer. Simultaneously, the control system issues a warning signal, requiring the activation of the external electrochemical energy storage unit.

[0078] The priority order of dynamic power allocation is as follows: the first priority is the furnace during the heating period S2 and the PCM melting rate is <0.7; the second priority is the furnace during the heat preservation period S4 and the PCM melting rate is >0.3; the third priority is the furnace during the coal charging period S1 and the coking period S5; and the fourth priority is the furnace during the constant temperature period S3.

[0079] The power allocation weighting formula is:

[0080] In the formula: For the first i The weighting of each furnace is as follows: furnace S2 w=4, furnace S4 w=2, furnaces S1 and S5 w=1, furnace S3 w=0.2.

[0081] The present invention also provides a dual-layer green electricity fluctuation buffer system based on phase change energy storage and multi-furnace coordination to realize the above-mentioned process, including a green electricity power prediction unit, a dual-layer coordination controller, an electrothermal power distribution unit, a multi-furnace coke oven body and an external electrochemical energy storage unit.

[0082] The green power prediction unit receives historical power data, weather forecast data, and power grid dispatch plan data at its input end. After processing the received data, it outputs the power prediction curve for the future period to the two-layer coordination controller.

[0083] The first input of the dual-layer coordinated controller is connected to the output of the green electricity power prediction unit to obtain the power prediction curve. The second input is connected to temperature sensors located at the interface of the phase change thermal storage layer in each furnace wall to obtain the real-time thermal state signal of the phase change thermal storage layer. The third input is connected to the operating condition monitoring module of each carbonization chamber to obtain the operating status signal of each furnace. The dual-layer coordinated controller runs a prediction, dual-layer scheduling, and response control algorithm. Based on the power prediction curve, the thermal state signal of the phase change thermal storage layer, and the operating status signal of each furnace, it generates power allocation commands for each carbonization chamber and outputs the power allocation commands to the electrothermal power allocation unit through the output terminal. At the same time, it outputs charge and discharge control commands to the external electrochemical energy storage unit. The sampling period of the dual-layer coordinated controller is 10 seconds, and the power regulation accuracy is ≤1%.

[0084] The power input terminal of the electric heating power distribution unit is connected to the green electricity input bus, and the control signal input terminal is connected to the power distribution command output terminal of the dual-layer coordination controller. The electric heating power distribution unit has multiple independent power channels, and the output terminal of each power channel is connected to the corresponding heating element in the carbonization chamber. Based on the received power distribution command, the electric heating power distribution unit distributes the electrical energy from the green electricity input bus to the heating elements in each carbonization chamber according to the set power of each channel. The rated total power of the electric heating power distribution unit is 36MW, the rated power of a single channel is 1MW, the response time is <100ms, and the distribution accuracy is <1%. The electric heating power distribution unit can receive power commands from the dual-layer coordination controller, adjust the power of each channel to the correct position within 100ms, and achieve independent adjustment of the power of each channel. There is no coupling interference between channels, and it can provide power limit protection, ensuring that the power of a single channel does not exceed 110% of the rated value, and recording the real-time value of the power of each channel.

[0085] The multi-furnace coke oven body comprises multiple carbonization chambers, each with a three-layer composite structure consisting of a refractory working layer, a phase change thermal storage layer, and a thermal insulation layer. The phase change thermal storage layer is divided into three temperature zones along the height of the carbonization chamber: a high-temperature zone, a medium-temperature zone, and a low-temperature zone. The high-temperature zone is filled with a sodium chloride-sodium sulfate mixed salt system, the medium-temperature zone with an aluminum-silicon eutectic alloy, and the low-temperature zone with a sodium nitrate-potassium nitrate binary molten salt. The phase change thermal storage layer provides isothermal buffering for power fluctuations from the electrothermal power distribution unit to the heating elements through the endothermic melting of the phase change material and the exothermic solidification. The thickness of the phase change thermal storage layer is 30–80 mm, and the heat storage capacity of a single furnace PCM is 150–350 kWh.

[0086] The control signal input terminal of the external electrochemical energy storage unit is connected to the charge / discharge control command output terminal of the dual-layer coordinated controller, and the power terminal is connected to the green electricity input bus. The external electrochemical energy storage unit charges or discharges according to the received charge / discharge control commands, supplementing and buffering power shortages or excesses outside the total power regulation range of the furnace group. The capacity of the external electrochemical energy storage unit is 5% to 10% of the total rated power of the furnace group.

[0087] The temperature monitoring system is equipped with one thermocouple at each interface between the refractory working layer and the PCM layer and the insulation layer in each carbonization chamber. The temperature measurement range is 0–1800℃, with an accuracy of ±0.5℃. The temperature at the refractory working layer-PCM layer interface is used to determine whether the PCM is in the phase transition range, i.e., whether the interface temperature is close to T_m; the temperature at the PCM layer-insulation layer interface is used to monitor the temperature gradient of the PCM layer and determine whether the PCM has undergone sufficient phase transition.

[0088] The PCM melting rate is estimated indirectly using a method based on the equivalent heat capacity formula. The melting rate of PCM is estimated in real time by monitoring the temperature change rate and heat power input of the PCM layer. Specifically, the method involves: under known conditions... P heater and P loss Under these conditions, the thermal power balance equation of the PCM layer is used. Reverse push P PCM ; and then according to P PCM The ratio of this ratio to the maximum phase change heat power is used to estimate the current melting rate. The estimation accuracy is approximately ±10%, which meets engineering control requirements.

[0089] Example 1: Buffer time that a single furnace can support PCM layer thermal response time constant Example of calculating the total power adjustment range of the furnace group.

[0090] (1) Buffer time that a single furnace can support Calculation example.

[0091] The design calculation is based on a NaCl-based mixed salt in the high-temperature zone: (The calculation is as follows) r =2330 kg / m 3 L=372 kJ / kg =0.65. The area of ​​one side wall of the carbonization chamber is approximately 15 m². 2 PCM layer thickness 50 mm, volume =0.75 m 3 ,but ≈117 kWh. (If the rated power of a single furnace is...) =1000 kW, buffering 25% power fluctuation, =117 / (1000×0.25)=0.47 h≈28min.

[0092] (2) Thermal response time constant of PCM layer Calculation example.

[0093] Calculations using typical parameters: r =2330 kg / m 3 , =1200 J / (kg K), =0.05 m, =15 W / (m K). Substituting, we get: = 2330 × 1200 × 0.05 2 / (15×3.1416 2 ) ≈ 38 s. That is, the PCM layer can reach 90% thermal power response in about 38 seconds.

[0094] (3) Example of calculating the total power adjustment range of the furnace group.

[0095] by N This design analysis is based on a 36-furnace system. The distribution of furnace numbers in each state is assumed to be: S1 coal charging period, 6 furnaces; S2 heating period, 8 furnaces; S3 isothermal period, 16 furnaces; S4 heat preservation period, 4 furnaces; and S5 coking period, 2 furnaces. The contribution of the PCM layer is not considered, and the system-level upscaling capacity is... = (6×0.15+8×0.3+16×0.05+4×0.2+2×0.1)×1000 kW = 4.1 MW. Considering the contribution of the PCM layer, and assuming... The average is 0.6. =200 kW / furnace, the system-level up-adjustment capacity increases to = 4.1 + 36 × 0.6 × 0.2 = 8.42 MW, approximately 23% If more furnaces are in S2 mode and the PCM is within its optimal buffer zone, the system-level upscaling capacity can reach 12.6 MW, approximately 35%. .

[0096] Example 2: Thermal Design and System Configuration of PCM Thermal Storage Layer in a 36-Furnace System This embodiment uses a 36-chamber electric coke oven as the design object to detail the complete implementation process of the PCM thermal storage layer from design to integration. The system has a total rated power of 36 MW, a single chamber rated power of 1 MW, and is equipped with a 200 MW photovoltaic power station and a 100 MW wind farm, achieving fully green electric drive. The carbonization chamber dimensions are 350 mm wide × 4300 mm high × 15980 mm long, with a single side wall area of ​​approximately 15.4 m². 2 The following details the implementation steps from five aspects: furnace wall structure design, PCM selection and layout, heat storage calculation, electric heating power distribution, and system integration.

[0097] (1) Detailed design of the three-layer composite structure of the furnace wall The furnace wall design incorporates a PCM heat storage layer in the middle of the existing two-layer structure of refractory working layer and thermal insulation layer, forming a three-layer composite structure of refractory working layer – PCM heat storage layer – thermal insulation layer. The total thickness of the three layers remains unchanged at 350 mm, consistent with the original furnace wall, without occupying additional space.

[0098] The refractory working layer is located on the inner side of the furnace wall and is made of high-alumina refractory bricks with a thickness of 180 mm. The working temperature range of this material is 1000–1150℃, and its thermal conductivity at room temperature is approximately 1.5 W / (m²). The refractory bricks (K) bear the main functions of temperature drop and high-temperature protection. Standard wedge-shaped bricks are used for laying, with brick joint width ≤2 mm. Phosphate refractory mortar is used for bonding during construction to ensure structural stability at high temperatures. A key design consideration for the refractory working layer is establishing a sufficient temperature gradient between the PCM layer phase transformation temperature (approximately 600–850℃) and the carbonization chamber working temperature (approximately 1000℃) to protect the PCM layer from excessive temperature shocks.

[0099] The PCM (Polymerized Ceramic Mold) storage layer, 50 mm thick, is located between the refractory working layer and the thermal insulation layer. The PCM is fixed using a honeycomb ceramic carrier adsorption method. The honeycomb ceramic carrier serves four purposes: (a) providing ample, uniformly distributed PCM storage space; (b) enhancing the effective thermal conductivity of the PCM layer; (c) preventing leakage and flow of liquid PCM; and (d) providing mechanical support to resist volumetric stress during thermal cycling. The dimensions of the honeycomb ceramic carrier match the dimensions of the refractory bricks, allowing it to be directly embedded between the brick joints, with construction methods similar to traditional refractory brick masonry.

[0100] The thermal insulation layer is located on the outer side of the furnace wall and adopts a double-layer structure. The inner layer is ceramic fiber felt, and the outer layer is lightweight insulating brick, both with a thickness of 60 mm. The two layers together are 120 mm thick. The ceramic fiber felt has excellent flexibility and thermal shock resistance, which can buffer the compression of the insulating brick by the thermal expansion of the PCM layer; the lightweight insulating brick provides structural support and further insulation. The design value of the outer surface temperature of the thermal insulation layer is ≤80℃, which meets the safe operation requirements of the furnace shell steel plate.

[0101] (2) PCM zoning selection and material configuration Based on the temperature distribution characteristics along the height direction of the carbonization chamber, the PCM layer is divided into three temperature zones along the height direction, and each temperature zone uses PCM material that matches the working temperature.

[0102] The high-temperature zone is located in the lower third of the height, ranging from 0 to 1.43 m, with an operating temperature of 850–1000℃. A NaCl-Na₂SO₄ mixed salt system is selected. The designed mixing ratio is 70% NaCl molar fraction and 30% Na₂SO₄ molar fraction. The designed eutectic melting point is 848℃, the latent heat of eutectic crystal is 372 kJ / kg, and the density is 2330 kg / m³. 3The single-sided PCM volume in this temperature range is 15.4 m³. 2 ×(1 / 3)×0.05 m =0.257 m 3 The mass of a single-sided PCM is 2330 × 0.257 = 599 kg. NaCl and Na2SO4 are both bulk chemical products, with raw material costs of approximately 0.3 yuan / kg and 0.8 yuan / kg, respectively. The cost of the mixed salt is approximately 0.45 yuan / kg. Therefore, the material cost of a single-sided PCM in this temperature range is approximately 270 yuan.

[0103] The intermediate temperature range is the middle third of the height, ranging from 1.43 to 2.87 m, with an operating temperature of 600–850℃. An Al-12Si eutectic alloy is used. It has a melting point of 577℃, a latent heat of 510 kJ / kg, and a density of 2660 kg / m³. 3 Thermal conductivity 160 W / (m K). The volume of the PCM on one side in this temperature range is 0.257 m³. 3 The mass of a single-sided PCM is 2660 × 0.257 = 684 kg. Al-Si alloy is a mature industrial material with a cost of approximately 18 yuan / kg, resulting in a total material cost of approximately 12,312 yuan for a single-sided PCM in this temperature range. This temperature range represents the region with the highest heat flux in the PCM thermal storage layer, and the high thermal conductivity of Al-Si alloy is beneficial for rapid response to power fluctuations.

[0104] The low-temperature zone is the upper 1 / 3 of the height, 2.87–4.30 m, with an operating temperature of 400–600℃. A NaNO3-KNO3 binary molten salt is used, with a mass ratio of 60:40. It has a melting point of 222℃, a latent heat of 110 kJ / kg, and a density of 1730 kg / m³. 3 The single-sided PCM volume in this temperature range is 0.257 m³. 3 The mass of a single-sided PCM is 1730 × 0.257 = 445 kg. The material cost is approximately 1.5 yuan / kg, and the material cost of a single-sided PCM in this temperature range is approximately 668 yuan.

[0105] The total mass of PCM on one side of the three temperature zones is 599 + 684 + 445 = 1728 kg, and the total mass on both sides is 3456 kg ≈ 3.5t / furnace. The total PCM usage for 36 furnaces is approximately 126 t.

[0106] (3) PCM heat storage design calculation According to the formula Calculate the PCM heat storage capacity for each of the three temperature zones. Take the fill ratio... or fill = 0.65.

[0107] High temperature zone unilateral = 2330 × 0.257 × 372 × 0.65 / 3600 = 39.9 kWh, totaling 79.8 kWh on both sides. Mid-temperature zone, single side. = 2660 × 0.257 × 510 × 0.65 / 3600 = 62.5 kWh, totaling 125.0 kWh on both sides. Low-temperature zone, single side. = 1730×0.257×110×0.60 / 3600 = 8.8 kWh, or fill Taking 0.60, the total heat storage capacity of both sides is 17.6 kWh. Total PCM heat storage capacity of a single furnace. = 79.8+125.0+17.6 = 222.4 kWh.

[0108] Calculate the buffer time that a single furnace can support. Use the rated power of a single furnace. = 1000 kW, power fluctuation range Three typical values, 5%, 15%, and 25%, were selected respectively. When... α When =5%, = 222.4 / (1000×0.05) = 4.45 h ≈ 267 min; when α = 15%, = 222.4 / 150 = 1.48 h ≈ 89 min; when α = 25%, = 222.4 / 250 = 0.89 h ≈ 53 min.

[0109] The above calculations show that for small fluctuations of 5%, the PCM layer can independently buffer for more than 4 hours; for moderate fluctuations of 15%, it can buffer for about 1.5 hours; and for large fluctuations exceeding 25%, it can buffer for about 53 minutes. This buffering capacity is far greater than the fast response time of the PCM layer. t PCM (≈ 38 s), which means that the PCM layer can both respond quickly to power changes and absorb power fluctuations for a longer period of time.

[0110] When the calculated buffer time that a single furnace can support is approximately 53 minutes, for a green electricity power fluctuation lasting 40 minutes, the PCM thermal storage layer alone can handle the buffering, without needing to activate furnace group coordinated control. When the green electricity power fluctuation lasts for 70 minutes, the PCM thermal storage layer first undertakes the rapid buffering for the first 53 minutes, and the power fluctuation corresponding to the remaining duration is handled by furnace group coordinated control. Therefore, the buffer time that a single furnace can support can be used as a basis for determining whether furnace group coordinated control needs to be activated.

[0111] (4) Configuration of electric heating power distribution unit The electric heating power distribution unit adopts a 36-channel independent power design, with each channel corresponding to one carbonization chamber. The input side of the unit is connected to the green electricity input terminal, and the output side is connected to the heating elements of each carbonization chamber via 36 independent solid-state relays. The total rated power of the electric heating power distribution unit is 36 MW, with a single channel rated power of 1 MW, a response time of <100 ms, and a distribution accuracy of <1%.

[0112] The electrothermal power distribution unit can receive power commands from the dual-layer coordination controller, adjust the power of each channel to the correct position within 100 ms, and achieve independent adjustment of 36 channels of power. There is no coupling interference between channels. It can provide power upper limit protection, ensure that the power of a single channel does not exceed 110% of the rated value, and record the real-time value of each channel's power. The sampling period is 10 s, and the data retention period is 1 min, which is used for PCM thermal state estimation and operation recording.

[0113] (5) Key points of system integration Temperature monitoring: One thermocouple is installed at the interface between the refractory working layer and the PCM layer and the insulation layer of each carbonization chamber. The temperature measurement range is 0–1800℃, with an accuracy of ±0.5℃. The temperature at the refractory working layer–PCM layer interface is used to determine whether the PCM is in the phase transition range, i.e., whether the interface temperature is close to... T m The temperature at the PCM layer-insulation layer interface is used to monitor the temperature gradient of the PCM layer and determine whether the PCM has undergone sufficient phase transformation. A total of 72 thermocouples are used in the 36 furnace chambers, and the signals are transmitted to the dual-layer coordinated controller.

[0114] Since directly measuring the melting rate of PCM is technically challenging, this embodiment employs an indirect estimation method. Based on formula... The equivalent heat capacity method estimates the melting rate of PCM in real time by monitoring the temperature change rate and heat power input of the PCM layer. Specifically, the method involves... P heater and P loss Under the condition of, through formula Reverse push P PCM ; and then according to P PCM The ratio of this ratio to the maximum phase change heat power is used to estimate the current melting rate. The estimation accuracy is approximately ±10%, which meets engineering control requirements.

[0115] Example 2: Design Condition Analysis of Two-Layer Cooperative Operation This embodiment, based on the 36-furnace system configured in Embodiment 2, details the specific implementation process of the dual-layer collaborative operation strategy under different design conditions. The analysis process consists of four steps: design condition construction, VMD power decomposition, dual-layer collaborative operation process design, and theoretical comparison of three buffer schemes.

[0116] (1) Design working condition construction The design conditions are based on a cloudy summer weather scenario. Three typical power fluctuation events are selected to verify the effectiveness of the two-layer collaborative strategy. The event selection principle is to cover different fluctuation amplitudes (small / medium / large) and different fluctuation durations (short / medium / long) to test the response capability of the two-layer collaborative strategy under various operating conditions.

[0117] Event A: A small, rapid fluctuation caused by cloud cover. Photovoltaic power generation suddenly dropped from 28 MW to 18.8 MW within 8 minutes, then recovered. This event is characterized by large fluctuations but short duration (<10 minutes), and is a typical fast-component dominant event. This event primarily tests the rapid response capability of the PCM layer; it is expected that the PCM layer should bear most of the power shortfall.

[0118] Event B: A significant and sustained fluctuation caused by the passage of thick clouds. Photovoltaic power generation dropped from 36 MW to 16.9 MW within 25 minutes, recovering after another 25 minutes. This event is characterized by large fluctuation amplitude and long duration (>10 minutes), exhibiting significant rapid and slow components. This event tests the layered relay capability of the two-layer coordination. The PCM layer responds quickly to the initial impact, while the furnace group coordinates to bear the subsequent sustained fluctuations.

[0119] Event C: Complex fluctuations caused by intermittent cloud cover. Multiple fragmented clouds intermittently block the sunlight, causing photovoltaic output to fluctuate repeatedly within 20 minutes, with a peak-to-valley difference of ±25%. This event is characterized by high fluctuation frequency and repeated changes in direction, representing a comprehensive test of the dynamic adaptability of the control strategy.

[0120] (2) VMD power decomposition and component extraction Taking event B as an example, the implementation process of VMD power decomposition is explained in detail. Original power curve. The power deficit decreased from 36 MW to 16.9 MW, resulting in a total power shortfall of 19.1 MW, lasting for 25 minutes.

[0121] The VMD decomposition parameter is set to the number of decomposition layers. K =4, quadratic penalty factor alpha=2000, convergence tolerance tol=1e-6, initial center frequency uniformly distributed in the range of 0 to 0.01 Hz. After VMD decomposition, 4 IMF components and 1 residual trend term are obtained.

[0122] The center frequencies and periods of each IMF component are as follows: IMF1 center frequency f1 = 0.003 Hz, corresponding period T1 ≈ 5.6 min, is a fast component; IMF2 center frequency f2 = 0.0008 Hz, corresponding period T2 ≈ 21 min, is an intermediate component; IMF3 center frequency f3 = 0.0002 Hz, corresponding period T3 ≈ 83 min, is a slow component; IMF4 center frequency f4 < 0.0001 Hz, corresponding period T4 > 167 min, is a residual trend term.

[0123] The IMF components are classified according to a 10-minute dividing period. When T1≈5.6 min<10 min, IMF1 is classified as a fast component, IMF2 and IMF3 are classified as slow components (although IMF2 has a period of 21 min, its amplitude is small, so it is classified as a slow component for unified processing), and IMF4 is a residual term and is not included in the buffer calculation.

[0124] The power gap corresponding to each component is: fast component P = IMF1 amplitude × total gap ratio ≈ 40% × 19.1MW = 7.6 MW; slow component P = (IMF2+IMF3) magnitude × total gap ratio ≈ 60% × 19.1 MW = 11.5 MW.

[0125] The basis for assigning the fast component to the PCM layer is the thermal response time constant of the PCM layer. It takes approximately 38 seconds to respond to rapid power changes with 90% efficiency within 1–2 minutes, while the period of the rapid component is 5.6 minutes, which is significantly longer. Therefore, the PCM layer can fully track the changes in the fast component. The reason why the slow component is handled by the furnace group coordination is that the period of the slow component is >10min, which is much greater than the continuous buffering capacity of the PCM layer, requiring the furnace group coordination to provide continuous power regulation capability.

[0126] (3) Detailed design of the two-layer collaborative operation process Taking event B as an example, the detailed design of each stage of the dual-layer collaborative operation is presented. It is assumed that the initial state distribution of each furnace is as follows: S1 Coal charging period (6 furnaces), S2 Heating period (8 furnaces), S3 Constant temperature period (16 furnaces), S4 Heat preservation period (4 furnaces), and S5 Coking period (2 furnaces). The initial melting rate of the PCM layer averages 0.5, all within the optimal buffer zone.

[0127] Phase 1: t=0~2 min is the rapid response period for the PCM layer. After a power drop, the dual-layer coordination controller detects the power change within 10 seconds and initiates a response. In this first phase, the power shortfall is entirely covered by the PCM layer, while the furnace group coordination layer remains inactive. The response of the PCM layer follows the formula... Calculate, take =38 s. When t=38 s, =7.6×(1-e -1 =7.6 × 0.632 = 4.8 MW; when t = 76 s, =7.6×(1-e -2 =7.6 × 0.865 = 6.6 MW; when t = 114 s, =7.6×(1-e -3 =7.6 × 0.950 = 7.2 MW. That is, the PCM layer undertakes approximately 7.2 MW within 2 minutes, accounting for 95% of the rapid component of 7.6 MW, basically achieving complete coverage of the rapid component. During this period, the furnace group coordination layer completes status acquisition and power planning calculation, preparing for the second stage.

[0128] Phase Two: t=2~10 min is the dual-layer coordinated transition period. The PCM layer continues to bear the fluctuating portion of the rapid component, approximately 7.6 MW, while the furnace group coordination layer's start-up power is redistributed. (According to formula...) The weighting strategy calculates the power adjustment for each furnace. First, the weighting of each furnace is determined. Furnaces during the S2 heating period (8 furnaces, w=4) have a total weight of 32; furnaces during the S4 heat preservation period (4 furnaces, w=2) have a total weight of 8; furnaces during the S1 coal charging period (6 furnaces, w=1) have a total weight of 6; furnaces during the S3 constant temperature period (16 furnaces, w=0.2) have a total weight of 3.2; and furnaces during the S5 coking period (2 furnaces, w=1) have a total weight of 2. Total weight =51.2.

[0129] The total power shortfall that the furnace group coordination layer needs to bear is the slow component of 11.5 MW minus the portion still borne by the PCM layer. In actual calculations, the cumulative heat release of the PCM layer at t=10 min is approximately 7.6 MW × 10 min = 76 MW. min=1.27MWh, accounting for approximately 5.7% of the total heat storage of the PCM (222.4 kWh), indicating that the PCM layer still possesses a high buffering capacity. To simplify the analysis, it is assumed that the PCM layer continues to bear the entire rapid component (7.6 MW) during the period from t=2 to 10 min; and the furnace group coordination layer bears the entire slow component (11.5 MW).

[0130] The power adjustment for each furnace is calculated as follows: For furnace S2, the adjustment per furnace is 11.5 × 4 / 51.2 = 0.898 MW. However, the rated power of a single furnace is only 1 MW. Currently, furnace S2 is in its heating phase, with a power of approximately 0.8 MW. The power can be increased to 1.1 MW, resulting in an actual increase of 0.3 MW per furnace, for a total increase of 2.4 MW for all 8 furnaces. For furnace S4, the adjustment per furnace is 11.5 × 2 / 51.2 = 0.449 MW. Furnace S4 is in its heat preservation phase, with a power of approximately 0.6 MW. The power can be reduced to 0.3 MW, resulting in an actual reduction of 0.3 MW per furnace, for a total reduction of 1.2 MW for all 4 furnaces. For furnace S1, the adjustment per furnace is 11.5 × 1 / 51.2 = 0.225 MW, for a total increase of 1.35 MW for all 6 furnaces. The adjustment amount per furnace in S3 is 11.5 × 0.2 / 51.2 = 0.045 MW, and the total increase for all 16 furnaces is 0.72 MW. The adjustment amount per furnace in S5 is 11.5 × 1 / 51.2 = 0.225 MW, and the total increase for both furnaces is 0.45 MW.

[0131] The total regulation capacity of the furnace group coordination layer = 2.4 (increase in S2) + 1.2 (decrease in S4) + 1.35 (increase in S1) + 0.72 (increase in S3) + 0.45 (increase in S5) = 6.12 MW. Adding the 7.6 MW borne by the PCM layer, the two layers together bear 13.72 MW. The remaining gap = 19.1 - 13.72 = 5.38 MW is supplemented by external electrochemical energy storage units. The external electrochemical energy storage units are configured at 5% of their rated power = 1.8 MW. With the 5.38 MW gap, the external electrochemical energy storage units operate at full power; any remaining shortfall is supplemented by grid dispatch.

[0132] Phase 3: The period from t=10 to 20 min is the dominant phase for the furnace group. As the PCM layer continues to release heat, the melting rate gradually decreases, from 1.0 to approximately 0.5 and then to approximately 0.2. The power handling capacity of the PCM layer decreases from 7.6 MW to approximately 3.8 MW (t=10 min) and then to approximately 1.5 MW (t=20 min). The furnace group coordination layer needs to handle a larger power deficit, but due to the limited adjustment margin of each furnace... α i With limited constraints, some power shortfall must be supplemented by external energy storage and grid dispatch. By t=20 min, the PCM layer handles about 1.5 MW, the furnace group coordination layer handles about 6 MW, which is close to the maximum regulation capacity, external energy storage handles 1.8 MW, and grid dispatch handles about 9.8 MW.

[0133] Phase Four: t=20–25 min is the PCM recovery period and power recovery. After t=20 min, power recovery begins, and the control system switches to recovery mode. Priority is given to replenishing power to the PCM layer: the power from the S2 heating furnace is preferentially allocated to the relatively empty S4 furnace with a melting rate <0.3, allowing its melting rate to recover to 0.4–0.5. The PCM layer absorbs approximately 2 MW of excess power within 5 min, and its melting rate recovers from 0.2 to 0.4. After full power recovery, the system enters standby mode, awaiting the next fluctuation event.

[0134] (4) Theoretical comparison of the three schemes The temperature control effects of three buffering schemes under the same event B condition are theoretically estimated. The estimation method is based on a first-order hysteresis thermal model. ,in For effective heat capacity, For ambient temperature, is the thermal time constant.

[0135] Option 1 has no buffer. The furnace wall is a traditional structure without a PCM layer, lacking multi-furnace coordination and energy storage. The single-furnace thermal time constant is... Approximately 50 minutes. When the power suddenly drops by 53%, the carbonization chamber temperature decreases according to a first-order hysteresis law, reaching a new steady-state temperature after about 50 minutes, a decrease of about 150°C. During the 25-minute fluctuation period, the temperature drops by about 120°C, approximately 80% of the steady-state change, and the temperature fluctuation amplitude is far greater than the process allowable range. This scheme relies entirely on the furnace's own thermal inertia; under large fluctuations, the temperature severely exceeds the limit, and the coke quality cannot be guaranteed.

[0136] Option 2 involves multi-furnace coordination only, without a PCM layer. The furnace walls have a traditional structure. N The multi-furnace coordination system (36) uses external energy storage at 15% of rated power. While multi-furnace coordination can redistribute power within 10 minutes, the high thermal inertia of the furnaces in the initial 2 minutes results in no effective buffering, causing a rapid temperature drop of approximately 30–40°C. Subsequently, with the furnace group coordination activated, the temperature drop rate slows, ultimately fluctuating by approximately ±50–80°C within 25 minutes. This scheme utilizes the power adjustment margin of multiple furnaces to partially mitigate power fluctuations, but its excessive thermal inertia prevents effective temperature reduction during the initial rapid fluctuations.

[0137] Scheme 3 is the dual-layer collaborative buffer of this invention. The PCM layer responds quickly within the initial 2 minutes, that is... Within approximately 38 seconds, temperature fluctuations were controlled within ±5℃; the furnace group coordination gradually took over over the following 8 minutes to maintain temperature stability; external energy storage provided supplementation when there was a significant power shortage. Throughout the entire 25-minute event, temperature fluctuations were controlled within ±5℃, meeting the process requirement of ±10℃ temperature fluctuations. The core advantage of the dual-layer coordination lies in the rapid response of the PCM layer, which fills the initial 2-3 minute response gap of the furnace group coordination, avoiding a sudden drop in initial temperature.

[0138] In summary, from the perspective of temperature control accuracy, Scheme 3 is far superior to Schemes 1 and 2; in terms of response speed, the PCM layer response time of Scheme 3 is about 10 times faster than the furnace group coordinated response time of Scheme 2; and in terms of energy storage investment, Scheme 3 requires only 5% of the rated power for energy storage configuration, while Scheme 2 requires 15%. The dual-layer collaborative strategy has significant advantages in all three dimensions: temperature control accuracy, response speed, and investment cost.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination, characterized in that, Includes the following steps: S1. Embedding a PCM heat storage layer in the furnace wall of the coke oven carbonization chamber creates a three-layer composite structure consisting of a refractory working layer, a PCM heat storage layer, and a thermal insulation layer, thereby establishing the PCM heat storage capacity. With a single furnace, the buffer time can be supported Quantitative relationships; S2. The power prediction curve is obtained through the green electricity power prediction unit. The power prediction curve is decomposed into multiple intrinsic mode functions using the variational mode decomposition algorithm. Each intrinsic mode function is then divided into fast and slow components according to the period. The period boundary is based on the thermal response time constant of the PCM thermal storage layer in S1. It is confirmed that the period of the fast component is less than Set a multiplier, the period of the slow component is greater than or equal to Set the multiplier; after completing the decomposition of green electricity power fluctuations, use the single furnace buffer time calculated by S1 as the control strategy switching criterion. When the power fluctuation duration corresponding to the fast component does not exceed the single furnace buffer time, execute S3, and the PCM thermal storage layer independently completes the buffering; when the power fluctuation duration exceeds the single furnace buffer time, execute S4, and start the furnace group coordinated control. S3, Based on the PCM heat storage established in S1 And a single furnace can support buffer time Based on the phase change thermodynamic characteristics of the PCM thermal storage layer, a thermal power balance equation for the PCM layer is established. The rapid component is isothermally buffered on a scale of seconds to minutes by the endothermic melting or exothermic solidification of the phase change material. The power fluctuation duration corresponding to the rapid component does not exceed the buffer time that a single furnace can support, calculated by S1. ; S4. For the slow component identified in S2, based on the power regulation margin difference of different working states of multiple furnaces, establish a macroscopic slow fluctuation power matching model for the furnace group. Calculate the power regulation margin of each furnace according to the working state coefficient of each furnace and the thermal state coefficient of the PCM heat storage layer after phase change buffering. Then, coordinate the power allocation of the slow component at the minute to hour level through state priority. S5. Establish a dual-layer collaborative operation strategy and PCM thermal state-driven power allocation decision. Combine the PCM layer thermal state, remaining heat storage and remaining single furnace buffer time output by S3, and the power adjustment margin of each furnace and furnace working status calculated by S4 to perform dynamic power allocation.

2. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, In S1, the PCM thermal storage layer adopts a zoned selection strategy. Based on the temperature distribution characteristics of the carbonization chamber along the height direction, the PCM thermal storage layer is divided into three temperature zones, specifically including: In the high-temperature zone, with a temperature range of 800–1000 ℃, a NaCl-Na2SO4 mixed salt system is selected, with NaCl and Na2SO4 mixed in molar fractions of 60%–80% and 20%–40%, respectively. For the medium temperature range of 600–800 ℃, Al-Si eutectic alloy is selected. In the low-temperature zone, with a temperature range of 400–600 ℃, a binary molten salt of NaNO3-KNO3 is selected.

3. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, The quantitative relationship between PCM heat storage and buffer time includes PCM heat storage and the buffer time that a single furnace can support; the buffer time that a single furnace can support is used to characterize the ability of a single furnace to independently buffer green power fluctuations without activating furnace group coordinated control; when the duration of green power fluctuation does not exceed When the green power fluctuation lasts longer than a certain period, only the PCM thermal storage layer is activated for buffering; when the green power fluctuation lasts longer than a certain period, only the PCM thermal storage layer is activated for buffering. At that time, the coordinated control of the furnace group is activated; The formula for calculating the heat storage capacity of PCM is: In the formula: Store heat for PCM; ρ PCM density; For PCM layer volume; L The latent heat of PCM; Fill rate; Buffer time that a single furnace can support The calculation formula is: In the formula: Rated power for a single furnace chamber; This represents the proportion of power fluctuation amplitude that needs to be buffered.

4. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, In S2, the green power prediction unit adopts the LSTM-XGBoost combined model. The input variables of the green power prediction unit include historical power data, weather forecast data and power grid dispatch plan. The output of the green power prediction unit is the power prediction curve.

5. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, S3 specifically includes the following steps: Establish the thermal power balance equation for the PCM layer: In the formula: The effective heat capacity of the carbonization chamber; Input power to the heating element; This is due to heat loss; The heat absorption / release power of the PCM phase change; The phase change heat power of PCM is calculated using the equivalent heat capacity method, as shown in the following formula: In the formula: δ represents the specific heat capacity of the solid state; L represents the latent heat; δ is the Gaussian distribution function, with its peak value located at the phase transition temperature. ; When the green electricity power suddenly drops by ΔP, the PCM layer compensates through heat release during solidification, as shown in the following formula: In the formula: The thermal response time constant of the PCM layer; PCM layer thermal response time constant Determined by the thermal resistance and thermal capacity of the PCM layer: In the formula: Equivalent heat capacity; PCM layer thickness; It is the equivalent thermal conductivity.

6. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, In S4, the formula for calculating the power regulation margin of each furnace is as follows: Formula for calculating upward margin: Formula for calculating downward margin: In the formula: This refers to the working status coefficient; The thermal state coefficient of the PCM; and These represent the maximum heat power that the PCM layer can release and absorb, respectively.

7. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, In S5, the two-layer collaborative operation strategy is as follows: When the fluctuation range of green electricity power Δ P <5% or duration Δ t When the temperature is less than 2 minutes, it is completely absorbed by the phase change thermal storage layer; When 5%≤Δ P <25% and 2 min≤Δ t When the time is less than 10 minutes, the phase change thermal storage layer bears 60% to 80% of the fluctuation, and the remainder is absorbed by the furnace group in a coordinated manner. When Δ P ≥25% or Δ t When the time is ≥10 min, the phase change thermal storage layer undertakes the initial... The rapid component within a time period is followed by the slow component, which is handled by the furnace group in a coordinated manner. When all furnace PCM layers are in a thermally depleted state, the system degenerates into a pure multi-furnace coordination mode.

8. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in claim 1, characterized in that, In S5, the priority order of the dynamic power allocation is as follows: The first priority is the furnace during the heating period and the PCM melting rate is <0.7; the second priority is the furnace during the heat preservation period and the PCM melting rate is >0.3; the third priority is the furnace during the coal charging period and the coking period; and the fourth priority is the furnace during the constant temperature period.

9. The dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination according to claim 8, characterized in that, The power allocation weight formula for the dynamic power allocation is: In the formula: For the first i The weighting of each furnace chamber is as follows: w=4 for the furnace chamber during the heating period, w=2 for the furnace chamber during the heat preservation period, w=1 for the furnace chamber during the coal charging and coking periods, and w=0.2 for the furnace chamber during the constant temperature period.

10. A dual-layer green electricity fluctuation buffering system based on phase change energy storage and multi-furnace coordination, used to implement the dual-layer green electricity fluctuation buffering process based on phase change energy storage and multi-furnace coordination as described in any one of claims 1-9, characterized in that, include: The green power prediction unit receives historical power data, weather forecast data, and power grid dispatch plan data at its input end. After processing the received data, it outputs the power prediction curve for the future period to the two-layer coordination controller. The dual-layer coordination controller has a first input terminal connected to the output terminal of the green electricity power prediction unit to obtain the power prediction curve, a second input terminal connected to the temperature sensor arranged at the interface of the phase change thermal storage layer of each furnace wall to obtain the real-time thermal state signal of the phase change thermal storage layer, and a third input terminal connected to the operating condition monitoring module of each carbonization chamber to obtain the operating status signal of each furnace. The dual-layer coordination controller runs a prediction, dual-layer scheduling and response control algorithm to generate power distribution instructions for each carbonization chamber based on the power prediction curve, the thermal state signal of the phase change thermal storage layer and the operating status signal of each furnace, and outputs the power distribution instructions to the electrothermal power distribution unit through the output terminal, while simultaneously outputting charge and discharge control instructions to the external electrochemical energy storage unit. The electric heating power distribution unit has its power input terminal connected to the green electricity input bus and its control signal input terminal connected to the power distribution command output terminal of the dual-layer coordination controller. The electric heating power distribution unit has multiple independent power channels, and the output terminal of each power channel is connected to the electric heating element of the corresponding carbonization chamber. The electric heating power distribution unit distributes the electrical energy of the green electricity input bus to the electric heating element of each carbonization chamber according to the set power of each channel according to the received power distribution command. The multi-furnace coke oven body contains multiple carbonization chambers. The furnace wall of each carbonization chamber adopts a three-layer composite structure of refractory working layer, phase change heat storage layer and heat insulation layer. The phase change heat storage layer is divided into three temperature zones along the height of the carbonization chamber: high temperature zone, medium temperature zone and low temperature zone. The high temperature zone is filled with sodium chloride-sodium sulfate mixed salt system, the medium temperature zone is filled with aluminum-silicon eutectic alloy, and the low temperature zone is filled with sodium nitrate-potassium nitrate binary molten salt. The phase change heat storage layer isothermally buffers the power fluctuations output from the electric heating power distribution unit to the electric heating element through the melting heat absorption or solidification heat release of the phase change material. The external electrochemical energy storage unit has its control signal input terminal connected to the charge and discharge control command output terminal of the dual-layer coordination controller, and its power terminal connected to the green electricity input bus. The external electrochemical energy storage unit charges or discharges according to the received charge and discharge control commands to supplement and buffer the power shortage or power excess outside the total power adjustment range of the furnace group.

Images