A clustering-based method for controlling room temperature and humidity uniformity in bamboo and wood chip drying

Abstract

The present invention discloses a method for controlling temperature and humidity uniformity in a bamboo and wood chip drying room based on clustering, and relates to the field of temperature and humidity control in a drying room. At present, there is a problem of uneven temperature and humidity in the drying room. The present invention comprises the steps of: performing clustering local abnormal factor detection on the pre-processed data to obtain the abnormal score of the temperature and humidity clustering cluster; analyzing the space with low temperature or high humidity according to the detection result of the clustered local abnormal factor, and locating the temperature and humidity uneven area; calculating the distance between the temperature and humidity uneven area in the X and Z axis planes and each air inlet, increasing the minimum distance outlet wind speed; expanding the air outlet at the X-axis in the area with low temperature or high humidity. The technical solution analyzes the space with uneven temperature and humidity, thereby improving the sampling accuracy of the temperature and humidity in the drying room, and achieves uniform temperature and humidity in the drying room by collaboratively adjusting the wind speed of multiple air inlets and outlets.

Description

Technical Field

[0001] The invention relates to the field of drying room temperature and humidity control, and in particular to a clustering-based method for uniformly controlling the room temperature and humidity of bamboo and wood chips drying rooms. Background Art

[0002] As people's living standards improve, environmental awareness is also constantly strengthened. Most families will choose solid wood materials for decoration. The demand for solid wood materials is increasing, and bamboo and wood drying technology has a huge impact on product quality. Uneven indoor temperature and humidity during the drying process causes bamboo and wood materials to crack, deform, dry unevenly, and even mold after drying. In response to this problem, the technology of uniform temperature and humidity in the drying room of bamboo and wood chips is studied to improve the yield rate of bamboo and wood material drying. However, the current control method is mostly PID control of single-point temperature and humidity acquisition, that is, to effectively control the temperature and humidity of the space near the temperature and humidity sensor, and the drying room space is large and complex. There is a problem of uneven temperature and humidity in the drying room, which causes the unevenly heated bamboo and wood materials to crack and deform, and even causes the bamboo and wood materials that have not reached the drying temperature and humidity for a long time to mold, thereby affecting the yield rate of the product. Summary of the invention

[0003] The technical problem to be solved and the technical task proposed by the present invention are to improve and perfect the existing technical solutions, and provide a method for controlling the temperature and humidity uniformity of a bamboo and wood chip drying room based on clustering, so as to obtain a bamboo and wood chip drying room with uniform temperature and humidity. To this end, the present invention adopts the following technical solutions.

[0004] A method for controlling humidity uniformity in a bamboo and wood chip drying room based on clustering, wherein the bamboo and wood chip drying room comprises a high-temperature heat pump dryer, a multi-speed adjustable air inlet channel, a drying room, a three-dimensional temperature and humidity sensor array, a hot air recovery channel and a main control unit, wherein the high-temperature heat pump dryer, the multi-speed adjustable air inlet channel, the three-dimensional temperature and humidity sensor array and the hot air recovery channel are connected to the main control unit through a control bus, and the high-temperature heat pump dryer is connected to the drying room through the multi-speed adjustable air inlet channel and the hot air recovery channel to form an airflow circulation channel; the high-temperature heat pump dryer comprises: an evaporator, a compressor and a condenser, the working fluid steam of the evaporator recovers the waste heat in the drying room, the working fluid steam obtains high-temperature and high-pressure steam through the compressor and enters the condenser to obtain hot air again, and the temperature and humidity sensors are evenly distributed in the drying room; the method for controlling humidity uniformity in a bamboo and wood chip drying room comprises the following steps:

[0005] 1) The main control unit samples the three-dimensional temperature and humidity sensor signal in the drying room to obtain temperature and humidity data;

[0006] 2) Preprocess the temperature and humidity data;

[0007] 3) Perform clustering local anomaly factor detection on the preprocessed data to obtain the temperature and humidity cluster anomaly score;

[0008] 4) Based on the clustered local anomaly factor detection results, analyze the space with low temperature or high humidity and locate the area with uneven temperature and humidity;

[0009] 5) Calculate the distance between the temperature and humidity uneven areas on the X and Z axis planes and each air inlet, and increase the wind speed at the minimum distance outlet;

[0010] 6) Expand the air outlet on the X-axis in areas with low temperature or high humidity.

[0011] As a preferred technical means: in step 3), the clustered local abnormal factor detection includes the following steps:

[0012] 301) constructing a weight matrix according to the temperature and humidity samples;

[0013] 302) obtaining the degree matrix and similarity matrix of the temperature and humidity in the drying room through the temperature and humidity sample data, obtaining the Laplace matrix, and performing singular value decomposition on the Laplace matrix;

[0014] 303) Take the singular value vectors corresponding to the first K smallest eigenvalues ​​of the Laplace matrix;

[0015] 304) clustering the singular value vectors of the Laplace matrix of the temperature and humidity sample data;

[0016] 305) After clustering, distinguish the size of the new sample drying room temperature and humidity data cluster;

[0017] 306) Calculate the temperature and humidity cluster anomaly score.

[0018] As a preferred technical means: in step 302), the degree matrix calculation formula of the temperature and humidity in the drying room is:

[0019]

[0020] where d i Composition degree matrix D, ω ij is an element of the similarity matrix W, and n is a natural number.

[0021] As a preferred technical means: in step 302), the similarity matrix calculation formula of the temperature and humidity in the drying room is:

[0022]

[0023] Where σ is the field width of the sample point, m i -m j It represents the Euclidean distance between any two points in the drying room temperature and humidity sample, sij Constitute the similarity matrix W.

[0024] As a preferred technical means: in step 302), the Laplace matrix L is calculated as follows: L=DW, where D is the degree matrix and W is the similarity matrix.

[0025] In step 303), the characteristic column vectors of the k temperature data and humidity data eigenvalues ​​are formed into a matrix U t ={u1,u2,...,u k},U h ={u1,u2,...,u k}, and let y i It's U t and U h The vector of the i-th row is used to obtain a new sample point of the temperature and humidity in the drying room; in step 304), the new samples of the temperature and humidity in the drying room are clustered into clusters C by using the K-means algorithm t1 ,C t2 ,...,C tk In step 305), cluster C h1 ,C h2 ,...,C hk Sort the elements from small to large, with b as the boundary between large and small clusters.

[0026] As a preferred technical means: in step 305), the size of the clusters is distinguished by sorting the temperature and humidity clusters in the drying room from small to large elements and using b as the boundary of the size clusters, wherein the size cluster boundary b is determined as follows:

[0027]

[0028] Among them, α represents the ratio of the data volume of the large cluster to the total data volume, LC represents the large cluster, SC represents the small cluster, β represents the data volume of the previous cluster is a multiple of the data volume of the next cluster, C represents the new temperature and humidity samples after K-means clustering, and b is used to distinguish large clusters from small clusters.

[0029] As a preferred technical means: when the temperature and humidity sample cluster data point of the drying room belongs to a large cluster, calculate its distance to the cluster center; when the temperature and humidity sample cluster data point of the drying room belongs to a small cluster, calculate its distance to the cluster center of the nearest large cluster, and obtain the abnormal score. The abnormal scores are sorted to obtain the temperature and humidity uneven space in the drying room; the temperature and humidity abnormal score of the drying room is calculated as follows:

[0030]

[0031] Where CBLOF(t) represents the abnormal score value of the tth temperature and humidity sample cluster.

[0032] As a preferred technical means: in step 4), the main control unit locates the uneven spatial coordinates (x b ,y b ,z b ).

[0033] As a preferred technical means: in step 5), the distance between the x-axis and z-axis coordinates of the drying room temperature and humidity uneven space and the x-axis and z-axis of each air inlet channel is calculated, and the calculation formula is:

[0034]

[0035] where x a ,z a is the x and z axis coordinates of each air inlet, x b ,z b are the x and z axis coordinates of the non-uniform space;

[0036] The wind speed of the air inlet channel with the smallest adjustment distance s is determined by the y-axis of the temperature and humidity uneven space in the drying room, and the calculation formula is:

[0037] sp=c*y b ,

[0038] Where sp is the air velocity at the outlet of the air inlet channel, c is the wind speed coefficient, y b is the y coordinate of the temperature and humidity non-uniform space.

[0039] As a preferred technical means: in step 6), the orientation of the hot air recovery channel is adjusted so that the recovery channel at the x-axis coordinate of the uneven space is enlarged to achieve uniform temperature and humidity in the drying room.

[0040] Beneficial effects:

[0041] (1) The method for uniform temperature and humidity in a bamboo and wood drying room established by the present invention analyzes the temperature and humidity non-uniform space based on three-dimensional temperature and humidity data using a clustered local anomaly factor detection algorithm, thereby improving the sampling accuracy of the temperature and humidity in the drying room.

[0042] (2) The present invention obtains equally divided spaces with uneven temperature and humidity by analyzing the three-dimensional temperature and humidity data in the drying room, and achieves uniform temperature and humidity in the drying room by collaboratively adjusting the wind speeds of multiple air inlets and outlets. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a structural schematic diagram of the present invention.

[0044] Figure 2 It is a coordinate definition diagram of the drying chamber of the present invention.

[0045] Figure 3It is the layout diagram of the temperature and humidity sensor of the present invention.

[0046] Figure 4 Structural diagram of the high-temperature heat pump dryer of the present invention.

[0047] Figure 5 Control flow chart of the present invention.

[0048] In the figure: 1. Drying room; 2. High-temperature heat pump dryer; 3. Air inlet channel; 4. Hot air recovery channel; 5. Temperature and humidity sensor; 6. Drying rack. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0050] like Figure 1 , 3 4, the bamboo chip drying room includes a high-temperature heat pump dryer 2, a multi-speed adjustable air inlet channel 3, a drying room 1, a three-dimensional temperature and humidity sensor 5 array, a hot air recovery channel 4 and a main control unit. The high-temperature heat pump dryer 2, the multi-speed adjustable air inlet channel 3, the three-dimensional temperature and humidity sensor 5 array, and the hot air recovery channel 4 are connected to the main control unit through a control bus. The high-temperature heat pump dryer 2 is connected to the drying room 1 through the multi-speed adjustable air inlet channel 3 and the hot air recovery channel 4, respectively. Figure 4 The working fluid steam of the evaporator in the high temperature heat pump dryer 2 recovers the waste heat in the drying chamber 1, and the working fluid steam is compressed by the compressor to obtain high temperature and high pressure steam and enters the condenser to obtain hot air again. Figure 3 The temperature and humidity sensors 5 are evenly distributed in the drying chamber 1; a plurality of drying racks 6 are provided in the drying chamber 1, and the drying racks 6 are divided into multiple layers, and each layer is provided with a plurality of temperature and humidity sensors 5, thereby forming a three-dimensional array of temperature and humidity sensors 5.

[0051] like Figure 5 As shown, the method for controlling the uniformity of room temperature and humidity in the drying of bamboo and wood chips comprises the following steps:

[0052] 1) The main control unit samples the three-dimensional temperature and humidity sensor signal in the drying room to obtain temperature and humidity data;

[0053] 2) Preprocess the temperature and humidity data;

[0054] 3) Perform clustering local anomaly factor detection on the preprocessed data to obtain the temperature and humidity cluster anomaly score;

[0055] 301) constructing a weight matrix according to the temperature and humidity samples;

[0056] 302) obtaining the degree matrix and similarity matrix of the temperature and humidity in the drying room through the temperature and humidity sample data, obtaining the Laplace matrix, and performing singular value decomposition on the Laplace matrix;

[0057] The three-dimensional temperature and humidity data samples X={t1,t2,....,t n}, Y={h1,h2,....,h n} Calculate the temperature and humidity similarity matrix W and degree matrix D in the drying room, and get the graph Laplace matrix L through W and D. The calculation formula is: L = DW, sort the eigenvalues ​​from small to large, take the first k eigenvalues ​​of temperature data and humidity data, and calculate the eigenvectors u1, u2, ...., u of the k eigenvalues ​​of temperature data and humidity data k ,v1,v2,....,v k ; The method for determining the temperature and humidity similarity matrix W in the drying room is as follows:

[0058]

[0059] Where σ is the field width of the sample point, m i -m j 2 represents the Euclidean distance between any two points in the drying room temperature or humidity sample, s ij The similarity matrix W is composed; the degree matrix D is the sum of the elements in each row of the drying room temperature and humidity similarity matrix W. The determination method is as follows: where d i Composition degree matrix D;

[0060] 303) Take the singular value vectors corresponding to the first K smallest eigenvalues ​​of the Laplace matrix;

[0061] The characteristic column vectors of the k temperature data and humidity data eigenvalues ​​are combined into a matrix U t ={u1,u2,...,u k},U h ={u1,u2,...,u k}, and let y i It's U t and U h The vector of the i-th row of is used to obtain the new sample point of temperature and humidity in the drying room;

[0062] 304) clustering the singular value vectors of the Laplace matrix of the temperature and humidity sample data;

[0063] By using the K-means algorithm, the new samples of drying room temperature and humidity are clustered into cluster C. t1 ,C t2 ,...,C tk ;

[0064] 305) After clustering, distinguish the size of the new sample drying room temperature and humidity data cluster;

[0065] Cluster C h1 ,C h2 ,...,C hk Sort the elements from small to large, with b as the boundary between large and small clusters; the boundary b between large and small clusters is determined as follows:

[0066]

[0067] Among them, α represents the ratio of the data volume of the large cluster to the total data volume, C is the clustering cluster of new samples of temperature or humidity in the drying room, LC represents the large cluster, SC represents the small cluster, β represents that the data volume of the previous cluster is a multiple of the data volume of the next cluster, and b is used to distinguish large clusters from small clusters.

[0068] 306) Calculate the temperature and humidity cluster anomaly score.

[0069] When the temperature and humidity sample cluster data point in the drying room belongs to a large cluster, calculate its distance to the cluster center. When the temperature and humidity sample cluster data point in the drying room belongs to a small cluster, calculate its distance to the cluster center of the nearest large cluster to obtain the abnormal score. Then sort the abnormal scores to obtain the temperature and humidity uneven space in the drying room. The temperature and humidity abnormal score is determined as follows:

[0070]

[0071] Where CBLOF(t) represents the abnormal score value of the tth temperature and humidity sample cluster.

[0072] 4) Based on the clustered local anomaly factor detection results, analyze the space with low temperature or high humidity and locate the area with uneven temperature and humidity;

[0073] According to the abnormal score value of the drying room humidity sample cluster cluster, the uneven spatial coordinates (x b ,y b ,z b ), the coordinates are defined as in the attached Figure 2 As shown,

[0074] 5) Calculate the distance between the temperature and humidity uneven areas on the X and Z axis planes and each air inlet, and increase the wind speed at the minimum distance outlet; the calculation formula is: where x a ,z a is the x and z axis coordinates of each air inlet, x b ,z bThe x and z axis coordinates of the uneven space are adjusted. The wind speed of the air inlet channel with the smallest distance s is determined by the y axis of the uneven temperature and humidity space in the drying room. The calculation formula is: sp = c*y b , where sp is the air velocity at the outlet of the air inlet channel, c is the wind speed coefficient, and y b is the y coordinate of the temperature and humidity non-uniform space,

[0075] 6) Expand the air outlet on the X-axis in areas with low temperature or high humidity;

[0076] Adjust the orientation of the hot air recovery channel to increase the size of the recovery channel at the x-axis coordinate of the uneven space to achieve uniform temperature and humidity in the drying room.

[0077] like Figure 2 As shown, the coordinate definition of the drying room takes the x, z, and y coordinates of 4 meters, 4 meters, and 12 meters respectively, and the x and z axis coordinates of the air inlet are (1,1), (3,1), (1,3), and (3,3) respectively.

[0078] The corresponding steps for controlling the room temperature and humidity uniformity of bamboo and wood chips drying are as follows:

[0079] First, the data collected by the temperature and humidity sensor array are preprocessed, and the clustered local anomaly point detection algorithm is used to obtain the anomaly score of each temperature and humidity cluster, which is sorted from large to small to obtain the temperature and humidity uneven area. The x, y, and z coordinates of the area are located as (1.1, 5, 0.9) as an example. The distances between the x, z axis coordinates of the drying room temperature and humidity uneven space and the x, z axis of each air inlet channel are calculated to be 0.141421, 1.902630, 2.10238, and 2.83196, respectively. The air inlet coordinate is (1,1) and the minimum distance is s min =0.141421, adjust the wind speed of the air inlet controller of the air inlet channel with coordinate (1,1) to 5*c, and the hot air recovery controller adjusts the position of the hot air recovery channel to increase the recovery channel at the x-axis coordinate 1, so that the hot air flow to the space with uneven temperature and humidity, that is, the coordinate (1.1, 5, 0.9), increases to achieve uniform temperature and humidity in the drying room.

[0080] The above-mentioned method for uniformly controlling room temperature and humidity during the drying of bamboo and wood chips based on clustering is a specific embodiment of the present invention, which has embodied the substantial characteristics and progress of the present invention. According to actual use needs and under the guidance of the present invention, equivalent modifications in shape, structure, etc. can be made to the method, which are all within the protection scope of this scheme.

Claims

1. A clustering-based method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips, characterized in that: The bamboo chip drying chamber comprises a high-temperature heat pump drying machine (2), a multi-speed adjustable air inlet channel (3), a drying chamber (1), a three-dimensional temperature and humidity sensor (5) array, a hot air recovery channel (4) and a main control unit. The high-temperature heat pump drying machine (2), the multi-speed adjustable air inlet channel (3), the three-dimensional temperature and humidity sensor (5) array and the hot air recovery channel (4) are connected to the main control unit via a control bus. The high-temperature heat pump drying machine (2) is connected to the drying chamber (1) via the multi-speed adjustable air inlet channel (3) and the hot air recovery channel (4) to form an air circulation channel. The high-temperature heat pump drying machine (2) comprises an evaporator, a compressor and a condenser. The working fluid steam of the evaporator recovers the waste heat in the drying chamber (1). The working fluid steam is compressed by the compressor to obtain high-temperature and high-pressure steam and enters the condenser to obtain hot air again. The temperature and humidity sensors (5) are evenly distributed in the drying chamber (1). The temperature and humidity uniformity control method for the bamboo chip drying chamber comprises the following steps: 1) The main control unit samples the three-dimensional temperature and humidity sensor signals in the drying room to obtain temperature and humidity data; 2) Preprocess the temperature and humidity data; 3) Perform clustering of local anomaly factors on the preprocessed data to obtain the temperature and humidity cluster anomaly scores; 4) Based on the clustered local anomaly factor detection results, analyze the space with low temperature or high humidity, and locate the area with uneven temperature and humidity; 5) Calculate the distance between the temperature and humidity uneven areas on the X and Z axis planes and each air inlet, and increase the wind speed at the minimum distance outlet; 6) Expand the air outlet on the X-axis in areas with low temperature or high humidity; In step 3), the clustered local anomaly factor detection includes the following steps: 301) Construct a weight matrix based on temperature and humidity samples; 302) Obtain the degree matrix and similarity matrix of the temperature and humidity in the drying room through the temperature and humidity sample data, obtain the Laplace matrix, and perform singular value decomposition on the Laplace matrix; 303) Take the singular value vectors corresponding to the first K smallest eigenvalues ​​of the Laplace matrix; 304) Clustering the singular value vectors of the Laplacian matrix of the temperature and humidity sample data; 305) After clustering, distinguish the size of the new sample drying room temperature and humidity data cluster; 306) Calculate the temperature and humidity cluster anomaly score.

2. The method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips based on clustering according to claim 1, characterized in that: In step 302), the degree matrix calculation formula of the temperature and humidity in the drying room is: ， in The composition degree matrix D, is the similarity matrix elements of , where n is a natural number.

3. The method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips based on clustering according to claim 2, characterized in that: In step 302), the similarity matrix calculation formula of the temperature and humidity in the drying room is: ， in is the area width of the sample points, It represents the Euclidean distance between any two points in the drying room temperature and humidity sample. Composition similarity matrix .

4. The method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips based on clustering according to claim 3 is characterized in that: In step 302), the Laplacian matrix The calculation formula is: ,in is the degree matrix, is the similarity matrix.

5. The clustering-based method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips according to claim 4, characterized in that: In step 305), the size of the clusters is distinguished by sorting the temperature and humidity clusters in the drying room from small to large and is the boundary between large and small clusters, where is determined as follows: in Indicates the ratio of the amount of data in a large cluster to the total amount of data. LC represents a large cluster, and SC represents a small cluster. Indicates that the amount of data in the previous cluster is a multiple of the amount of data in the next cluster. represents the new temperature and humidity samples after K-means clustering. Used to distinguish large clusters from small clusters.

6. The clustering-based method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips according to claim 5, characterized in that: When the temperature and humidity sample cluster data point of the drying room belongs to a large cluster, calculate its distance to the cluster center. When the temperature and humidity sample cluster data point of the drying room belongs to a small cluster, calculate its distance to the cluster center of the nearest large cluster to obtain the abnormal score. Then sort the abnormal scores to obtain the temperature and humidity uneven space of the drying room. The calculation formula of the temperature and humidity abnormal score of the drying room is as follows: in Indicates The abnormal score value of the temperature and humidity sample cluster.

7. The clustering-based method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips according to claim 6, characterized in that: In step 4), the main control unit locates the uneven spatial coordinates according to the abnormal score values ​​of the drying room humidity sample clusters .

8. The clustering-based method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips according to claim 7, characterized in that: In step 5), calculate the drying room humidity uneven space Axis coordinates and air inlet channels The distance between the axes is calculated as: ， in For each air inlet Axis coordinates, For uneven space Axis coordinates; Adjust distance The wind speed of the minimum air inlet channel is determined by the uneven temperature and humidity in the drying room. The axis is determined by the calculation formula: ， in is the air velocity out of the air inlet channel, is the wind speed coefficient, For spaces with uneven temperature and humidity coordinate.

9. The clustering-based method for controlling the uniformity of room temperature and humidity in drying bamboo and wood chips according to claim 8, characterized in that: In step 6), adjust the hot air recovery channel orientation to make the uneven space The recovery channel at the axis coordinate is enlarged to achieve uniform temperature and humidity in the drying room.

Images