Plasma processing apparatus and prediction method of the condition of plasma processing apparatus
A system that predicts an apparatus state of a plasma processing apparatus including a processing chamber in which a sample is processed is configured to have a data recording unit that records emission data of plasma during processing of the sample and electrical signal data obtained from the apparatus during the plasma processing, an arithmetic unit that includes a first calculation unit for calculating a first soundness index value of the plasma processing apparatus and a first threshold for an abnormality determination using a first algorithm with respect to the recorded emission data and a second calculation unit for calculating a second soundness index value of the plasma processing apparatus and a second threshold for the abnormality determination using a second algorithm with respect to the electrical signal data recorded in the data recording unit, and a determination unit that determines soundness of the plasma processing apparatus using the calculated first soundness index value and the first threshold and the calculated second soundness index value and the second threshold.
1. A plasma processing apparatus in which a state is predicted by an analysis apparatus, comprising:
a processing chamber in which a sample is subjected to plasma processing,
wherein the analysis apparatus predicts the state based on a first soundness index value obtained by analyzing a first data obtained from the plasma processing apparatus using a first algorithm and a second soundness index value obtained by analyzing a second data obtained from the plasma processing apparatus using a second algorithm,
wherein the first algorithm is an algorithm in which a first time of the plasma processing is a time in which the state is determined as abnormal by the first time soundness index value, and
wherein the second algorithm is an algorithm in which a second time of the plasma processing is a time within a range of a third time to a fourth time,
wherein the second time of the plasma processing is a time in which the state is determined as abnormal by the second soundness index value,
wherein the third time is a time obtained by subtracting a predetermined time from the first time, and
wherein the fourth time is a time obtained by adding the predetermined time to the first time.
2. The plasma processing apparatus according to claim 1 ,
wherein the first data is emission data obtained from plasma during the plasma processing, and
wherein the second data is data different from the first data.
3. The plasma processing apparatus according to claim 2 ,
wherein principal component analysis (PCA) is used as the first algorithm, and
wherein auto associative kernel regression (AAKR) is used as the second algorithm.
4. The plasma processing apparatus according to claim 2 ,
wherein the analysis apparatus removes a radio frequency component from the second data, and then analyzes the radio frequency component removed second data by using the second algorithm.