Load estimating device for rolling bearing, control device for mechanical device provided with rolling bearing, load estimating method, and program

A load estimating device for a rolling bearing, comprising:

• a vibration sensor configured to measure vibration of the rolling bearing during rotation;
• a rotational speed sensor configured to measure a rotational speed of the rolling bearing during rotation;
• a derivation portion configured to derive a vibration value of a predetermined vibration frequency using vibration information measured by the vibration sensor;
• an estimation portion configured to estimate a load applied to the rolling bearing, the load corresponding to the vibration value of the predetermined vibration frequency derived by the derivation portion and the rotational speed measured by the rotational speed sensor, using a learned model obtained by performing learning processing using data including a pair of the load applied to the rolling bearing as well as the vibration value of the predetermined vibration frequency and the rotational speed of the rolling bearing as learning data and using the load applied to the rolling bearing as output data;
• a determination portion configured to determine a vibration frequency of interest from a theoretical frequency of the rolling bearing and a high-order vibration frequency thereof based on the rotational speed measured by the rotational speed sensor;
• and
• a control portion configured to control at least one of torque around a shaft supported by the rolling bearing and rotation of the rolling bearing in accordance with the load estimated by the estimation portion, wherein
• the derivation portion is configured to derive a vibration value of the vibration frequency determined by the determination portion.

A load estimating method for a rolling bearing, comprising:

• a first measurement step of measuring vibration of the rolling bearing during rotation;
• a second measurement step of measuring a rotational speed of the rolling bearing during rotation;
• a derivation step of deriving a vibration value of a predetermined vibration frequency using vibration information measured in the first measurement step;
• an estimation step of estimating a load applied to the rolling bearing, the load corresponding to the rotational speed measured in the second measurement step and the vibration value derived in the derivation step, using a table in which a correspondence relationship among the load applied to the rolling bearing, the vibration value of the predetermined vibration frequency, and the rotational speed is defined;
• a determination step of determining a vibration frequency of interest from a theoretical frequency of the rolling bearing and a high-order vibration frequency thereof based on the rotational speed measured in the second measurement step;
• and
• a control step of controlling at least one of torque around a shaft supported by the rolling bearing and rotation of the rolling bearing in accordance with the load estimated by the estimation step, wherein
• in the derivation step, a vibration value of the vibration frequency determined in the determination step is derived.

A load estimating method for a rolling bearing, comprising:

• a first measurement step of measuring vibration of the rolling bearing during rotation;
• a second measurement step of measuring a rotational speed of the rolling bearing during rotation;
• a derivation step of deriving a vibration value of a predetermined vibration frequency using vibration information measured in the first measurement step;
• and
• an estimation step of estimating a load applied to the rolling bearing, the load corresponding to the vibration value of the predetermined vibration frequency derived in the derivation step and the rotational speed measured in the second measurement step, using a learned model obtained by performing learning processing using data including a pair of the load applied to the rolling bearing as well as the vibration value of the predetermined vibration frequency and the rotational speed of the rolling bearing as learning data and using the load applied to the rolling bearing as output data;
• a determination step of determining a vibration frequency of interest from a theoretical frequency of the rolling bearing and a high-order vibration frequency thereof based on the rotational speed measured in the second measurement step;
• and
• a control step of controlling at least one of torque around a shaft supported by the rolling bearing and rotation of the rolling bearing in accordance with the load estimated by the estimation step, wherein
• in the derivation step, a vibration value of the vibration frequency determined in the determination step is derived.

A non-transitory computer-readable storage medium storing a computer program configured to cause a computer to function as:

• a first acquisition portion configured to acquire vibration information of a rolling bearing during rotation;
• a second acquisition portion configured to acquire a rotational speed of the rolling bearing during rotation;
• a derivation portion configured to derive a vibration value of a predetermined vibration frequency using the vibration information;
• and
• an estimation portion configured to estimate a load applied to the rolling bearing, the load corresponding to the rotational speed acquired by the second acquisition portion and the vibration value derived by the derivation portion, using a table in which a correspondence relationship among the load applied to the rolling bearing, the vibration value of the predetermined vibration frequency, and the rotational speed is defined;
• a determination portion configured to determine a vibration frequency of interest from a theoretical frequency of the rolling bearing and a high-order vibration frequency thereof based on the rotational speed measured by the rotational speed sensor;
• and
• a control portion configured to control at least one of torque around a shaft supported by the rolling bearing and rotation of the rolling bearing in accordance with the load estimated by the estimation portion, wherein
• the derivation portion is configured to derive a vibration value of the vibration frequency determined by the determination portion.

A non-transitory computer-readable storage medium storing a computer program configured to cause a computer to function as:

• a first acquisition portion configured to acquire vibration information of a rolling bearing during rotation;
• a second acquisition portion configured to acquire information on a rotational speed of the rolling bearing during rotation;
• a derivation portion configured to derive a vibration value of a predetermined vibration frequency using the vibration information;
• and
• an estimation portion configured to estimate a load applied to the rolling bearing, the load corresponding to the vibration value of the predetermined vibration frequency derived by the derivation portion and the rotational speed acquired by the second acquisition portion, using a learned model obtained by performing learning processing using data including a pair of the load applied to the rolling bearing as well as the vibration value of the predetermined vibration frequency and the rotational speed of the rolling bearing as learning data and using the load applied to the rolling bearing as output data;
• a determination portion configured to determine a vibration frequency of interest from a theoretical frequency of the rolling bearing and a high-order vibration frequency thereof based on the rotational speed measured by the rotational speed sensor;
• and
• a control portion configured to control at least one of torque around a shaft supported by the rolling bearing and rotation of the rolling bearing in accordance with the load estimated by the estimation portion, wherein
• the derivation portion is configured to derive a vibration value of the vibration frequency determined by the determination portion.